Wednesday, December 10, 2003


9:10 a.m.











Hotel Washington

515 Fifteenth Street, N.W.

Washington, D.C.




LINDA REED, Acting Executive Secretary































    DARYL B. LUND, Ph.D.















Welcome and Introductions:

    Dr. Sanford Miller    5


Conflict of Interest Statement:

    Ms. Linda Reed    10


Opening Remarks:

    Dr. Robert Buchanan    13


Subcommittee Reports    17


     Infant Formula: Ms. Jeanne Latham

    Questions of Clarification    19

    Public Comment

      Roger Clemens, Dr.PH    21

    FAC Discussion and Comment    39


     Contaminants and Natural Toxicants,

     Enterobacter sakazakii:  Dr. Frank Busta,

    Ms. Jeanne Latham    44

    Questions of Clarification    48

    Public Comment

      William C. MacLean, Jr., M.D.    58

    FAC Discussion and Comment


More Opening Comments:

    Joseph A. Levitt    69

    Dr. Robert Brackett    81


More Subcommittee Reports


     Dietary Supplements: Dr. Johanna Dwyer,

    Ms. Jeanne Latham    83

    Questions of Clarification    87


     Additives and Ingredients: Dr. Johanna Dwyer,

    Mr. Richard Bonnette    97

    Questions of Clarification    102


     Food Biotechnology:  Dr. Frank Busta,

    Dr. Michael Watson    110

    Questions of Clarification    114





Status Report and Response to Food Advisory

Committee's Recommendations on Methylmercury

in Fish and Shellfish    129


FDA and EPA Development of Joint Advisory:

    Dr. David Acheson, CFSAN    129

    Ms. Denise Keehner, EPA    135

    Dr. David Acheson, CFSAN    142

    Questions of Clarification    157


Exposure Assessment and Peer Review:


     Peer Review: Rita Schoeny    177

     Exposure Assessment:  Dr. Michael Bolger    188

                           Dr. Clark Carrington

    Questions of Clarification    198


Welcome and Introductions

    DR. MILLER:  Good morning.  I would like to take this opportunity to thank you all for coming today, the members of the advisory committee, of which I think there are some eight new members.

    We have several topics to be discussed today including reports of several subcommittees that function in relationship to the senior committee, of which this is it. There are a couple of things.  First, I intend to be ruthless in terms of time.  I have said this the last time the committee met and I will say it again; there are too many people on the agenda and, if we don't be rigorous in terms of the amount of time that each person uses, we will be here next Tuesday.

    I think that is not what all of us want.  I have a suspicion that, by tomorrow afternoon, people are going to be sneaking out and that wouldn't be fair to the people that still have time to make presentations.  So I will use the gavel,

the presentation gavel, which is an inside joke, ruthlessly.  So, please, I really beg of you, please stick to your times.

    Second of all, the function of these meetings is to inform the committee.  So the only questions that will be asked will be asked by members of the committee.  The audience will not be called upon to provide questions to members of the committee or people making presentations.  So, only the committee will be called upon to ask questions, if they have them.

    I also ask the committee if, when you do ask a question, let's keep the questions as brief as we can make them but at least make them as informative as possible because the function of the committee is going to be to come up with a report for CFSAN based on the outcome of this committee.

    Thirdly, one of the topics being discussed today is infant formula.  There are two of us that have worked in this area before with industry and we have recused ourselves for these issues.  I have asked Bob Russell to chair because I am one of the

ones that is recused.  Doug Archer is the other one.

    I think that is all of the announcements.

    Before we begin, I wonder if I could take a few minutes here and have the committee introduce themselves.  Let's start with Dr. Scherer.

    DR. SCHERER:  Clifford Scherer, Cornell University.

    MS. HALLORAN:  I am Jean Halloran from Consumers Union.

    DR. ACHOLONU:  Alex Acholonu from Alcorn State University, Mississippi.

    DR. APOSHIAN:  Vas Aposhian, University of Arizona.

    DR. WASLIEN:  Carol Waslien, University of Hawaii.

    DR. DOWNER:  Goulda Downer, Metroplex Health and Nutrition Services here in Washington, D.C.

    DR. KRINSKY:  Norman Krinsky, Tufts University,  School of Medicine, Boston.

    DR. CALLERY:  Patrick Callery, West

Virginia University.

    DR. RUSSELL:  Robert Russell, Nutrition Research Center, Tufts, Boston.

    MS. REED:  Linda Reed, Center for Center for Food Safety and Applied Nutrition.

    DR. MILLER:  Sandy Miller, Virginia Tech University.

    DR. LUND:  Daryl Lund, University of Wisconsin, Madison.

    DR. DICKINSON:  Annette Dickinson, Council for Responsible Nutrition.

    DR. ARCHER:  Doug Archer, University of Florida.

    DR. BUSTA:  Frank Busta, University of Minnesota.

    DR. DURST:  Richard Durst, Cornell University.

    DR. ALLER:  Marion Aller, Florida Department of Agriculture and Consumer Services.

    DR. NELSON:  Mark Nelson, Grocery Manufacturers of America.

    DR. DWYER:  Johanna Dwyer, Tufts New

England Medical Center.

    DR. HEIMBURGER:  Doug Heimburger, University of Alabama at Birmingham.

    DR. MILLER:  Thank you.  There is one more thing that I think I would rather say now than later.  This is the last meeting that Catherine DeRoever, who has acted as the Executive Secretary of this committee--I was going to say almost since its inception.  I think it was since its inception.  So, to a large extent, the committee and the way it works has been the result of Catherine's kind but firm directions, particularly the way she controls the Chairman.  It always ends up the way she wants it to be, somehow, and everybody agrees that is the way it should have been in the first place.

    In any case, this is her last meeting and I would like to offer to her the thanks of the committee for doing a sensation job.  Her successor, Linda Reed, whom you have already met,I am sure will be a worthy successor.

    Thank you, Catherine.


    Slight change on the agenda.  Joe Levitt, the Director of the Center, who also is retiring--I am just wondering if I have anything to do with it.  We need to think about that--will not join us until just before lunch.  He will have with him the new Director of the Center to meet the committee and to be introduced to the committee.

    Bob Buchanan of the Center will make some opening remarks about where CFSAN would like us to go on these discussions.  So, the first item on the agenda today is the additional Conflict of Interest Statement.  Linda Reed will provide that.

Conflict of Interest Statement

    MS. REED:  Good morning, as Chairman Miller indicated, I am Linda Reed, the Acting Executive Secretary for this Food Advisory Committee Meeting.  I would like to welcome everyone, our members and our temporary voting members.  Thank you very much for being here today.

    With that, I would like to read the conflict of interest statement into the meeting record.  The authority to appoint temporary voting

members is granted to the Center Director.  Relying on that authority, Mr. Levitt, Director, Center for Food Safety and Applied Nutrition, has signed letters stating, by the authority granted under the Food Advisory Committee Charter, I hereby appoint Dr. Aposhian, Dr. Russell and Dr. Scherer as temporary voting members of the Food Advisory Committee for December 10 and 11, 2003, committee meeting.

    All members and temporary voting members have been screened for financial conflicts of interest.  Because of the breadth of topics to be discussed at this meeting, all of the members and temporary voting members have been screened for any and all financial interest associated with the regulated industry.

    Based on this review, FDA has determined, in accordance with 18.USC Section 208(b)(3), to grant general-matters waivers to Dr. Marion Aller, Dr. Douglas Archer, Dr. Francis Busta, Dr. Richard Durst, Dr. Johanna Dwyer, Dr. Douglas Heimburger, Dr. Normal Krinsky, Dr. Sanford Miller, and Dr.

Carol Waslien that permits them to participate fully in the matters before this committee.

    Copies of the waiver statements may be obtained by submitting a written request to the Agency's Freedom of Information Office, Room

12A-30, of the Parklawn Building.  However, as Dr. Miller stated, he requests that I inform the committee that he is voluntarily recusing himself from all discussions relating to the Infant Formula Committee.

    Similarly, Dr. Douglas Archer has asked that I inform the committee that he is also recusing himself from all discussions relating to the Contaminants and Natural Toxicants Subcommittee on Enterobacter sakazakii.

    In an effort to enhance consistency within FDA, the agency has recently adopted a policy whereby all public commenters will be asked to report any personal financial interests that could be affected by the committee's deliberations.  A copy of this policy was provided to all the individuals who registered to make a public

comment.  I believe that additional copies of the policy may be obtained from the registration desk.

    I have one additional administrative announcement.  We have received two written submissions for individuals that were not able to appear at the meeting.  The submissions are from Dr. Jane Hightower and Dr. Philippe Grandjean.

The submissions have been provided to our members and copies are available at the registration desk.

    With that, I would like to welcome everyone and turn the meeting back over to Dr. Miller.

    DR. MILLER:  Thank you, Linda.

    The next item on the agenda is some opening remarks which will be made by Bob Buchanan.

Opening Remarks

    DR. BUCHANAN:  Thank you, Sandy.  As Sandy already sort of let the cat out of the bag, I am not Mr. Levitt.  Joe sends his regrets for not being able to be here first thing in the morning.  He does intend to be here just before the noon break and will be making a few comments at that


    I did want to take this time to just make a couple of opening remarks and sort of to set the stage for the next deliberations that will be taking place.  I want to emphasize the fact that FDA faces challenges in the area of foods and applied nutrition that are diverse, often highly complex and affect the American consumers, every American consumer every day.

    FDA prides itself on basing its programs and its decisions and continually striving to bring the best science it can to these questions and working through a process of effective risk management.  An integral part of being able to achieve that goal of sound science is seeking opinions and advice and expertise from outside.

    We do this in many manners.  We deal effectively with our stakeholders.  We establish collaborative research and scientific programs.  As you know, being here today, we rely heavily on our advisory committee.

    We will be throwing to you today some very

highly complex issues and we appreciate, and I want to thank you ahead of time for the attention and efforts that you are going to bring to bear.

    I might note also that is a

precedent-setting meeting because this is the first meeting we have had where we have completed on full cycle of our new structure of the parent committee and six standing subcommittees.  You will be hearing reports from those subcommittees this morning.

    I want to note also that the committee has been rechartered for its next two-year cycle and we are looking forward to your activities over that next two years.  I was also asked to let everyone know that there are still several vacancies available on our subcommittees.  So, as you are listening to the activities of this morning of those subcommittees, if you have suggestions on people that would be appropriate and interested in serving in that capacity, please let Linda Reed know.  That also goes for the people in the audience, or pass the word back to other interested

parties.  We are looking for the best scientists that can be brought to bear in those areas.

    With that, again, I want to express our sincere thanks for devoting your time and efforts and substantial grey matter to the issues at hand and, if there is anything that we can do to make your activities easier during the next several days, please do not hesitate to ask the staff to identify how we can help you through this process.

    With that, I also want to express my sincere thanks to Sandy for his continuing activities and also to Cathy DeRoever for her activities on behalf of the committee for the last nine years.

    So let's get on with the show.  Thank you.

    DR. MILLER:  Thank you, Bob.

    The next item on the agenda has to do with the Subcommittee Report on Infant Formula.  Since we need to make a changeover in the Chairmanship, we will take a really quick five-minute break and reassemble for this discussion.


Subcommittee Reports

Infant Formula

    DR. RUSSELL:  We are at the point of hearing the first subcommittee report, the Report on Infant Formula.  To give that report is Ms. Jeanne Latham who is the Executive Secretary of Infant Formula and Dietary Supplements Subcommittees.

    DR. LATHAM:  Good morning, everyone.  I am Jeanne Latham, the Executive Secretary for the Infant Formula Subcommittee.  I am standing in for Dr. Cutberto Garza who chaired the November 2002 Food Advisory Meeting on Infant Formula.  I will very, very briefly review the minutes of the meeting.

    In your briefing package for today's meeting, you received, among other things, a copy of these minutes for the November 18 to 10, 2002, Food Advisory Committee Meeting on Infant Formula.

    As a matter of background, the Food Advisory Committee met in April of 2002 to begin discussions related to the general question, what

components constitute an appropriate and complete general science-based set of guiding principles for clinical studies used in the context of providing assurances that a particular infant formula supports normal physical growth under its intended conditions of use.

    The November 18 to 19, 2002 meeting was a follow-up to that April, 2002 meeting and addressed questions relating to that general question.  The purpose of the November, 2002 meeting was to consider seven questions under two issues; criteria for the adequate evaluation of normal physical growth during the first six months as an indicator of the nutritional adequacy of new infant formulas and the types of changes in infant formulas that should be accompanied by a clinical study in order to provide assurances of normal physical growth.

    Questions 1 through 6--again, these are in your minutes in your package--address the specific criteria covered under the first issue about criteria for evaluating normal physical growth and the types of changes in infant formulas that should

be accompanied by the clinical study were addressed in Question 7.

    The committee reached consensus on all seven questions and made recommendations to the agency.  I was not going to go into the details because you have that before you and, if you had any comments or questions, I will try my best to respond to you.

Questions of Clarification

    DR. RUSSELL:  This is the time for questions of clarification from the Food Advisory Committee members if you have any questions at that point.

Public Comment

    If there are no questions from the committee of Ms. Latham, we will go ahead with regard to the public comment on the Infant Formula Report.  Before we do that, I need to read to you this statement that, both the Food and Drug Administration and the public believe in a transparent process for information-gathering and decision-making.

    To ensure such transparency at the Open Public Hearing Session of the Advisory Committee Meeting, FDA believes that it is important to understand the context of an individual'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 relationship that you may have with the sponsor, its product and, if known, its direct competitors.

    For example, this financial information may include the sponsor's payment of your travel, lodging or other expenses in connection with your attendance at the meeting.  Likewise, the Food and Drug Administration encourages you, at the beginning of your statement, to advise the committee if you do not 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, however.

    The first speaker is Roger Clemens from

the University of Southern California who is the Industry Rep on the Infant Formula Subcommittee.

    You have eight minutes, Dr. Clemens.

    DR. CLEMENS:  Thank you, Mr. Chairman.  Roger Clemens with the University of Southern California School of Pharmacy and the Infant Formula Council, International Infant Formula Council, has paid my way here to serve as consultant on this topic.

    On behalf of the International Infant Formula Council and its membership, I would like to thank you for this opportunity to address the question posed to this advisory committee just moments ago by Ms. Latham.

    Historically, researchers in the

infant-formula industry have conducted clinical studies in infants prior to the introduction of a new formula or whenever an existing formula underwent a change that might call into question the formula's continued ability to support growth.  In the process, researchers follow ICH and American Academy of Pediatric Guidelines.

    It should be emphasized that, during the past ten years, approximately 50 growth studies among 6,000 infants have been performed for new infant-formula introductions characterized by what the FDA and the Academy have termed as major change in formulation.  Based on these practices, no infant formula has been introduced into the marketplace that has been later withdrawn due to failure to support normal growth of infants.  This confirms that the clinical process currently followed in assessing growth adequacy are practical and appropriate.

    We believe there are several areas, general areas, of the subcommittee's charge that deserve particularly emphasis.  Changes are frequently made in the manufacture of infant formula in the interest of continuous improvement and not every change is a major change, and not every major change justifies a clinical study.

    Guidelines incorporated into law in the 1986 Amendments to the Infant Formula Act specify the types of changes that require clinical study.

These guidelines highlight the need for individual case assessment before making any decision to subject babies to clinical trials.  The recommendations before you are not final.  The agency has advised that the Infant Growth Report on your desk is, in fact, a work in progress and we look forward to continually working with the subcommittee in the future on these issues.

    Current systems are firmly in place by the industry and they are working.  Practical and definitive outcomes need to be defined for any growth studies when considered necessary.  Certainly, there is little point in having collected data unless there is a recognized measure by which to evaluate those data.

    Industry experience in this arena is rich and valuable to this subcommittee and to these deliberations.  Even industry assistance in framing questions for future FAC meetings should be very helpful to the FDA in eliciting answers that provide truly practical guidance for the agency's review process.

    Each of you has this flow diagram in the packet before you.  Some of you have seen this industry-provided decision tree before, such as the last meeting.  It is key to understanding the process that industry routinely goes through in determining the nature of a given change and whether, in light of the nature of the change, clinical data are necessary to confirm growth support.

    This decision tree reflects a process that has served the public well.  This FAC has been asked by the FDA to assess and respond to seven fundamental questions.  Upon review of the FAC responses, we offer the following summary comments.

    Formulation changes; it is clear that not all change trigger the need for a clinical study.  As indicated in the decision tree, there are multiple factors to be considered when documenting the nutritional adequacy of infant formula.  In many cases, a clinical-growth study may be warranted.  In other cases, not.  For example, when clinical growth studies are not justified include

taking into account previously published information.

    Control population; the reported FAC response to this key topic is there should be a comparison in addition to some reference source based on breast-fed infants to try to understand deviations, if any, between growth patterns exhibited by breast-fed infants and formula-fed infants targeted for the study.

    Breast-fed infants are not mentioned in the question posed to the FAC.  Breast-fed infants do not represent a concurrent control and there is not a well-recognized growth database for

breast-fed infants.

    Clinical comparators; if a reference is used to help interpret the results of a randomized controlled parallel-designed clinical study, the CDC-NCHS reference data published in 2000 reflect the most recent information and should serve as the comparator for these kinds of clinical studies.

    Essential anthropometrics; at issue is the need for anthropometric assessment from birth to

six months of age for infants implying that this need applies to studies for both term and pre-term infants.  Historically, for term infants, the Academy recommended that growth studies went from birth to four months of age.  There is no rationale for an extra two months of follow up, particularly when extending the growth evaluation beyond four months of age is confounded by the introduction of solid foods.

    With respect to preterm infants, formula for this population is provided during the initial hospitalization after birth and is usually discontinued post-discharge.  Therefore, to demonstrate adequate growth in preterm infants, the study period can be no longer than from near the time that enteral feedings were introduced until the approximate hospital discharge.

    Value of metrics; given the circumstances surrounding anthropometric measure obtained at birth, trace measurements are not reliable baseline measurements.  Recording body weight and recumbent length data provides sufficient data for


    Reference groups; in evaluating normal physical growth for term and preterm infants, the FAC noted the comparison should be the population intended to consume the formula.  We do not believe that the subcommittee's comments were meant, or should be interpreted to mean, to preclude conducting growth studies outside of the United States.  What is most important is the health status of the intended population and not its geographical location.

    Growth assessment endpoints; consistent with our experience and the extensive data collected and published, body weight, alone, is the most reproducible and the most reliable measure to assess body growth and, thus, the most important factor to evaluate.

    In conclusion, our experience and data reveal that the current clinical practices in assessing adequacy to support growth are practical and appropriate.  Therefore, any proposed changes to these practices must be carefully considered and

evaluated in the context of whether they would truly improve the quality of infant formula.

    Thank you for your time.

    DR. RUSSELL:  Thank you very much.  Are there any questions at this time from the Food Advisory Committee members?  Dr. Dwyer?

    DR. DWYER:  Roger, I wasn't sure what your point was on breast milk.

    DR. CLEMENS:  Thank you, Johanna, for the comment.  It was thought that we should run a concurrent "control" to gain additional data and our experience has indicated that we know--I believe it was the FAC had indicated that they wanted additional data on the deviations or any changes in kids that were breast-fed.

    In fact, we know quite well that the growth velocities of breast-fed kids do differ from those which are formula-fed.  We know that the growth patterns are not identical between these two groups.  It has been well established.  Despite these apparent differences, there isn't any information whatsoever to say that these

differences are clinically significant.

    MS. HALLORAN:  Did the committee give any consideration to indicators considering developmental indicators as opposed to just physical growth and weight gain?

    DR. CLEMENS:  I defer that question back to the committee.

    DR. RUSSELL:  Can you repeat the question, Ms. Halloran?

    MS. HALLORAN:  The question was was there any consideration given to whether one should look at developmental indicators as well as physical growth.

    DR. RUSSELL:  It would have to be referred back to the subcommittee?

    DR. CLEMENS:  It was deferred.  The point was discussed, but no consensus was established.

    DR. RUSSELL:  Other questions or clarification or discussion?

    DR. CLEMENS:  Thank you.

    DR. RUSSELL:  Thank you very much, Dr. Clemens.

    The next speaker is Ms. Sally Fallon, Westin Price Foundation in Washington who has requested fifteen minutes.

    MS. FALLON:  Thank you very much for allowing me to present this morning.  My name is Sally Fallon.  I am the President of the Westin A. Price Foundation, a nonprofit nutrition education foundation based in Washington, D.C.  We, myself and members of the Board of Directors of this foundation, have no ties with any food industry, certainly not with the infant-formula industry.  We serve as volunteers.

    Our concern is the use of soy infant formula.  About 25 percent of formula-fed babies in the United States receive soy infant formula.

    Our first concern centers on the major ingredient, soy protein isolate.  This is a highly processed protein powder derived from soy that is processed using many chemicals and at very high temperatures in order to reduce the levels of antinutrients that naturally occur in the soy bean.  There is always a tradeoff.  In order to reduce the

antinutrients as much as possible, there is also the overdenaturing of many of the proteins in soy.

    In feeding experiments with rats, soy protein isolate caused increased requirements for Vitamins E, K, D and B12 and deficiency symptoms of calcium, magnesium, manganese, molybdenum, copper, iron and zinc.

    As far as we can determine, soy protein isolate does not have GRAS status and we have not been able to ascertain whether it has gotten premarket approval for soy infant formula.

    We are also concerned about toxins that are added or produced during processing, nitrates and lysinoalanine, which were the concerns expressed by the FDA in the early 1970s when GRAS status was sought for protein isolate, higher than normal levels of aluminum, fluoride and MSG, free glutamic acid.

    The second concern is the lack of cholesterol in soy infant formula.  Dietary cholesterol is vital for the development, optimal development, of the infant, particularly

neurological development.  This is why mother's milk is very rich in cholesterol and contains a special enzyme that ensures the complete assimilation of the cholesterol.  Soy infant formula does not contain cholesterol.

    Our third concern is the phytoestrogens, the high levels of phytoestrogens, in soy infant formula. These have been associated with many conditions, endocrine disruption, depression of the immune system and thyroid dysfunction.  Animals fed high levels of phytoestrogens have shown infertility, reproductive problems, thyroid disease and liver disease.  This has been observed in many species including mice, rats, cheetah, sturgeon, quail, sheep, pigs and marmoset monkeys.

    Children on soy-based formula have levels of estrogens 13,000 to 22,000 times higher than the levels in children on milk-based formula.

    This indicates a comparison.  We are looking here at the--on a body-weight basis, we have three surveys from Japan showing a range of 0.17 to 0.47 milligrams per kilogram of body weight

daily.  This is what you find in the Asian diet.  Getting above that at 0.60 to 0.75, you start to see thyroid problems, endocrine disruption, hormonal changes.  The FDA recommended amounts comes out to be about 0.42 milligrams per kilogram of body weight.

    An infant on soy infant formula because of its small size--and this is an average--is about 0.625 milligrams per kilogram of body weight, so this is an order of magnitudes greater than any thing that we see in traditional diets or that is causing problems in adults.  You have to remember the infant is particularly vulnerable to the effects of these isoflavins.

    Just to go over a few studies from the literature.  This is 1981, soy formula fed to premature babies caused increase in digestive enzymes indicating very low digestibility of soy infant formula.

    Gastrointestinal damage was observed in two infants--this was published in 1983--this was consistent with celiac disease and lectin-induced


    Aluminum, this was 1986, was found to be ten to twenty-fold greater in cow's mild formula compared to breast milk but a hundred-fold greater in the soy infant formula.

    The things that I am pointing out here would not necessarily affect growth.  In fact, the estrogens promote growth.  In animals, they give estrogens to promote growth.  So the protocols that you are suggesting, just six months looking at weight and height, are not going to see the neurological effects of some of these other components of soy formula.

    Fort, 1986, found that twice as many

soy-fed children developed diabetes as those in the control group that was breast-fed or received

milk-based formula.  Once again, this was not show up in the first six months of life.

    Early puberty in girls; this was the Puerto Rican study, 1986.  Soy infant formula was associated with higher rates of early development in girls including breast development and pubic

hair before the age of eight and sometimes before the age of three.

    1987, soy-based or milk-free formulas contain eight to fifteen times more cadmium than milk-based formulas as well as high amounts of fluoride.

    Dr. Fort, once again, 1990; he documented associative soy formula in infancy with autoimmune thyroid problems.  Once again, these types of problems would not necessarily show up in the first six months of life.

    The JAMA study, published in 2001, widely reported as a vindication of soy formula, actually found more reproductive problems and more asthma in the adults who had been fed soy formula.  It was also an indication of higher weight and thyroid problems.

    Now, this was a study on marmoset monkeys that is very important to our discussion, published in 2002.  They fed marmoset monkeys soy infant formula for the first few months of life.  There was a decrease in the testosterone surge of 70

percent and the levels of soy given to these monkeys were actually less than what is given to human infants on soy formula.

    As you may be aware, the testosterone surge in little boys in the first six months of life programs the child to express male characteristics at puberty.  So, once again, you are not going to see the effects of this until much later.

    Thymic and immune changes in mice due to exposure to genistein in utero or after birth; the authors caution--in the conclusion, these results raise the possibility that serum genistein concentrations found in soy-fed infants may be capable of producing thymic and immune abnormalities.

    This is another recent study, 2002.  Infant mice given genistein developed cancer of the uterus later in life.  They said the data suggest that genistein is carcinogenic if exposure occurs during critical periods in the young animal's development.

    Changes in the ovaries.  This was another recent study.  Once again, these are, I believe it is mice.  The authors concluded that, given that human infants are exposed to high levels of genistein in soy-based foods, this study indicates that the effects of such exposure on the developing reproductive tract warrant further investigation.

    Another study, quite recently, 2003, when female rats were given low, medium and high doses of genistein, they found persistent demasculinization of male infants even at the low doses.

    Finally, another one in 2003, infant rats exposed to phytoestrogens exhibited much higher levels of anxiety and stress and much less contact with the other animals.  This is consistent with the anecdotal reports that we get on a weekly basis of extreme emotional behavior and difficulty in social adaption on soy-fed infants.  Once again, this is not going to show up in looking at growth in the first six months of life.

    Our recommendations; we urge the FDA to

follow the cautionary principle and take soy formula off the market, make it available only by doctor's prescription and to provide pediatricians with the recent studies indicating the dangers of high levels of isoflavins in the diets of infants.

    We think that the promotion and development of a meat-based formula should be a top priority.  Gerber used to produce a meat-based formula.  Finally, we encourage renewed emphasis on breast feeding and healthy diets for breast-feeding mothers.

    We are a nonprofit nutrition education foundation.  We encourage you to visit our website.  We promote a return to traditional foods and nutrient-dense diets, especially for our pregnant women, nursing women and growing children.

    Thank you very much for the opportunity to testify today.

    DR. RUSSELL:  Thank you.  Your talk was not provided to the committee members in the briefing materials.  I wonder if you are planning to do that for us or if your talk is on your

website that you have posted up there.

    MS. FALLON:  I can get it on the website right away.  I do have notes here.  We also provided about 60 or 70 pages of testimony.  I don't know if it is in your packet, but we certainly submitted a lot of testimony.

    DR. RUSSELL:  It is not in the packet that this committee received.  But a copy of your talk would be very helpful.

    MS. FALLON:  I have it here.  I will leave it with Linda.

    DR. RUSSELL:  Thank you.  If you wouldn't mind staying up there for a little bit.  If the committee has any discussion or comment on this, this is time to ask questions or to get clarifications or comments.

FAC Discussion and Comment

    DR. RUSSELL:  Dr. Aller?

    DR. ALLER:  Not so much a question for our speaker but rather for the committee, or for the subcommittee, was this topic discussed at the subcommittee at all?  Again, I am just curious.

    MS. LATHAM:  We did not single out any specific types of--infant formulas were not singled out such as this.  They were groups of different types of formulas were discussed as a whole.

    MS. FALLON:  We did submit testimony to both committee meetings, both times.

    DR. RUSSELL:  Any other questions?

    DR. CALLERY:  This is Pat Callery.  A follow up of your question.  Did the subcommittee discuss developmental measures beyond just the weight and height that would be more in the neurological area and in the reproductive area that were described here?

    MS. LATHAM:  I believe we were just basically talking about the physical growth factors and we primarily talked about the anthropometrics.  I believe that there was some very brief discussion about neurologic, but that was the focus of the meeting.

    DR. RUSSELL:  Dr. Dwyer?

    DR. DWYER:  I haven't had a chance to read it yet, but the Committee on Nutrition of the

American Academy of Pediatrics periodically puts out a Yellow Book that reviews a lot of topics on infant feeding.  I don't remember the past Yellow Books singling this out after they have all of the looks at all the evidence.

    Could you, perhaps, tell us why you think that the studies you have suggested come to such a different conclusion?

    MS. FALLON:  I hate to speculate.  The concerns have been expressed by public groups only in the last few years.  We have actually been trying to get a meeting with the American Academy of Pediatrics.  Our phone calls are never returned.

    We were just founded in 1999 and kind of got going in 2000.  Our Director of Public Affairs has attended all of the meetings on infant formula and we have presented a large number of studies that we urge the committees to look at.  We would very much like to open a dialogue with the American Academy of Pediatrics.  Our Director of Public Affairs was in Chicago just recently but was unable to get a meeting.

    DR. RUSSELL:  Dr. Durst?

    DR. DURST:  I am not a nutritionist, but I think studies such as this point out the importance of not only considering growth but that long-term effects are very critical.  I can imagine there are a number of infant formulations that can provide growth but may actually have other deleterious effects that don't show up until later in life.

    DR. DOWNER:  Goulda Downer.  I think what we really need, though, would be to get a plethora of data to suggest what she has presented today.  Clearly, there may be a potential for long-term effects of not just soy but other products.  But just one or two pieces of research is not enough.  And I am a nutritionist.

    DR. RUSSELL:  I have a question.  Dr. Russell.  With regard to the JAMA study that you talked about, I actually forget whether--you mention the higher rates of the asthma and reproductive abnormalities in adults who have been fed soy infant formulas.  Were those results actually statistically significant?

    MS. FALLON:  No; they were not.  It was not a very large study, either.  It would have been very hard to get statistical significance.  There were many flaws in that study and I can send you an analysis that we did of the study.  Just the way the questions were asked and the parameters that they used, we felt that the study was very superficial and very flawed.

    I would be happy to provide you with our analysis of that.

    DR. RUSSELL:  I think that would be helpful, yes.  Thank you.

    Any other questions or comments?  If not, thank you very much.

    MS. FALLON:  Thank you.

    DR. RUSSELL:  At this point, while we reorganize again, another very short break of five minutes while Dr. Miller resumes the Chair here.  Thank you.


    DR. MILLER:  I want to thank Dr. Russell for taking over the Chair.  What I am going to do

is to ask certain members of the committee to take some notes on the discussion so that we will have some kind of record when the meeting is over to make some recommendations, if necessary, to CFSAN.

    The next topic is from the Contaminants and Natural Products Subcommittee which has to do with Enterobacter sakazakii.  The presenter will be Dr. Frank Busta who is the Chairman of the Committee.


Contaminants and Natural Toxicants

Enterobacter sakazakii

    DR. BUSTA:  Thank you, Sandy.  As you heard, I am Frank Busta.  The committee had a very interesting and comprehensive meeting.  As you can see in the report, we had a series of presentations from guest speakers including the marketing and use of powdered infant formula, a review of the case of a Tennessee outbreak as well as case reports from British outbreak situations.

    Then the medical consequences of Enterobacter sakazakii, and I will just say E. sak

for all of our ease.  It is a lot faster.  Infections as well as the general microbiology, the detection methodology, thermal and other resistances by the organism, and then some field studies that FDA conducted.

    There were a series of public comments from a number of individuals representing industries as well as the International Formula Council and an individual from a medical college.

    As you can see in your report, the questions that we received as a charge was, one, characterize the infants at risk.  We were asked is there a risk and, if so, identify the infants that are at risk.  Here is a situation where we actually voted unanimously on the answer to that question and saying that, yes, there is a risk.

    We identified, and you can see the comprehensive identification of the infants that would be at risk, mainly the preemies, immunocompromised infants and those in neonatal intensive-care units.  The subcommittee thought that there was probably a low but yet unquantified

risk to healthy term infants and really couldn't describe that at the point.

    Our second charge was, if there is a meaningful risk, how can the risk be addressed.  As you see, the first question is intervention strategies in manufacturing.  We discussed that extensively, methods to reduce bacterial presence in the manufacturing process.  There is a series of interventions that were identified.

    The development of microbiological sampling programs was also discussed and it was indicated that FDA and the industry formally assess the contribution of a microbiological testing program, whether it would help with the intervention.

    In discussing the currently available processing technologies, it appears that E. sak cannot be completely eliminated for the at-risk populations.  We really don't know if the indicated interventions will actually improve the situation.

    Continuing on in other intervention strategies in handling practices and other aspects,

it was recommended that the FDA, with input from the industry, prepare educational documents and distribute those and get those to the healthcare users because it was apparently some healthcare users do not understand that powdered infant formula is not sterile and, consequently, there can be some mishandling and, if it is possible, based on available information, specify allowable lower limits for the microbial detection of E. sak.  The subcommittee said there was not sufficient information to make that kind of judgment at the present time.

    Finally, the subcommittee identified a series of information gaps, a lot of them ideal research and surveillance needs including inactivation post-drying of E. sak.  If I were to do this again, I would separate the first bullet into two.  The second one is continuing the methods to detect E. sak.

    Number three there is to document the occurrence of the organism in powdered infant formula, develop means of sterilizing powdered

infant formula, develop sterile liquid products for use in the at-risk populations, identify pathogenic factors in the organism that result in the disease, do continued population surveillance on the incidence of the infection, assure that

clinical-laboratory procedures are available to identify the organism and the last, but far from least, determine and develop the optimal therapy for infected infants.

    Thank you.

    DR. MILLER:  Thank you, Frank.

    Any questions from the committee?

    DR. DURST:  Frank, is there any information available on the resistance of E. sak to sterilization?  Is it especially robust or is it something that could be easily sterilizable.

    DR. BUSTA:  We obviously have one of the experts sitting over here to my right, but, as I recall, it appears that some of the E. sak is more resistant to dehydration, may have a fair amount of resistance, but the normal handling of

infant-formula components, it would appear that at

least some of the processes would inactivate the organism in it.  So, post-processing contamination appears to be one of the major sources.

    Bob, am I still on the right track?

    DR. BUCHANAN:  It is not a particularly heat-resistant organism but, within the Enterobacteriaceae, at least some of the strains have some fairly substantial heat resistance at non-boiling temperatures.  It does survive quite nicely in dehydrated products.  We have it surviving for well over a year now in infant formula.  It does also have the characteristic of coming out of the dry state and growing quite aggressively so it does represent some unique challenges.

    But, in the greater scheme of things; no, it is not a heat-resistant organism though it is relatively heat resistant to close cousins, E. coli, Salmonella, et cetera.

    DR. MILLER:  Would everybody identify themselves for the record, please.

    DR. BUCHANAN:  Sorry.  Bob Buchanan, FDA.

    DR. DURST:  That was Richard Durst asking the question.

    DR. MILLER:  Thank you.  Dr. Dwyer?

    DR. DWYER:  I was just wondering where it came from, and I think it is not just

post-production.  It is in the raw materials as well, or isn't that known?

    DR. BUSTA:  I am not sure that the comprehensive investigation of that is complete.  I think it has been found in the environment in the plant and its origin, I am not sure it has been highly traced as to the origin.  I am waiting to see--am I, again, interpreting that right?  I got a nod from Dr. Buchanan.

    But, as I recall, it has been identified in the processing-plant environments and, consequently, that is one of the interventions that is being worked on.

    DR. DWYER:  May I ask a follow-up question?

    DR. MILLER:  Of course.

    DR. BUSTA:  Yes.

    DR. DWYER:  If it is true that what you say is that every effort should be made to avoid feeding powdered infant formula to at-risk infants who include immunocompromised infants at any age, does that mean that--what do you do with powdered formula in Africa, for example, where are lot of very sick babies are being fed and I assume they are being fed powdered.

    DR. BUSTA:  That is probably a larger question than I can answer, but I know that, for clinical situations or neonatal situations where the infants are ill or preemies, it was suggested, I think by some of the committee, that they might just use sterilized liquid formula.  Then the concern was that the individualized needs for each of these situations is such that the production of sterile liquid infant formula for each of those groups is at least not yet feasible.

    There was also a concern about rehydrating with very hot water so that you would inactivate the organism with the heat treatment of just rehydrating.  The concern that I recall expressed

was hazardous situations of using boiling water to rehydrate the formula and possibly losing some of the nutrients in the rehydration of that hot water.

    For international use in immunocompromised populations such as in Africa, I don't recall that being surfaced at the meeting that we had.

    DR. MILLER:  Dr. Lund?

    DR. LUND:  I was going to ask a procedural question on this.  What does this committee do with the report from the subcommittee?  Do we take an action on this report?

    DR. MILLER:  The major function of this meeting is for the subcommittees to report back to the parent committee to provide information.  Our role is to ask questions that maybe should be rediscussed or relooked at, or not.  If we are satisfied with the report, then that is okay, too.

    DR. LUND:  I guess that leads to the question, Frank, there are recommendations in here with regard to specific things that should be done.  With regard to the information that Dr. Durst asked with regard to its susceptibility to various

treatments, et cetera, that is also one of the recommendations from this committee?

    DR. BUSTA:  Yes.  As I recall, the study of the organism and the study of interventions are, I think, ongoing but encouraged by the subcommittee.

    DR. MILLER:  Members of the committee ought to be prepared when we come to the end of the meeting to indicate where they disagree with the subcommittee reports.  I would like you to raise that question now, now that we are discussing it.  But, when all is said and done, the committee is going to have to sign their names to a report that says that we can find no area of disagreement, not because they are necessarily experts, but based on our general knowledge of the area.

    What I am also going to do is to ask some of you to serve as kind of a monitor and to prepare a brief note of discussion as it goes on, not of the report, itself, because that we have copies of, but a brief note of the discussion and any important points that came out of the discussion

that we think ought to be highlighted in the final report.

    DR. KRINSKY:  Krinsky, Tufts.  I just wanted to know if there was any follow up to the recommendation that the FDA should prepare educational documents.  Has that been carried out in the nine months since the subcommittee reported?

    MS. LATHAM:  This is Jeanne Latham.  That is something that is in process with that process.

    DR. MILLER:  Any other comments or questions?

    DR. DOWNER:  Yes.  Goulda Downer.  How long would something like that take before the public becomes aware of the recommendations, the educational materials?

    DR. MILLER:  How long does it take the recommendations of the committee?

    DR. DOWNER:  For the educational materials to be developed.  As I think about Johanna's question, and she talked about Africa, I am also thinking about, as the consumer rep, those resource-poor communities in this country and those

communities that have a lot of migrant populations that prefer to use the powdered formula than the liquid ones.

    So, based on that, I am asking how soon will those recommendations be available for the public?

    DR. MILLER:  Bob has an answer?

    DR. BUCHANAN:  This is Bob Buchanan from the FDA.  The questions that were posed to the subcommittee were specifically oriented to the situation here in the United States.  In addition, the Food and Drug Administration has brought this issue up through the Codex Alimentarius system and has requested that the Codex Alimentarius Food Hygiene Committee reexamine the current code of practice for powdered infant formulas.  This is an agenda item currently being considered at that level.

    In addition, at our request, we have gone to WHO-FAO to request an expert consultation on this subject matter looking not just at the situation here in the United States but infant

formula on a broader framework, or that is what Codex Food Hygiene would like.

    It is my understanding that that consultation workshop is going to take place in early February.  We have yet to receive the official notice, but I understand it is on its way out.  So this is being dealt with at two different levels but the specific request to the subcommittee was oriented towards the condition here in the United States.

    DR. MILLER:  Johanna?

    DR. DWYER:  Just to follow up again, Dr. Buchanan, to make your life more difficult, I guess both Dr. Downer and I were thinking about at-risk populations such as the infants of HIV-positive mothers who often are immunosuppressed.  It seems to me that any information activities would have to involve somehow getting information to those healthcare providers and, perhaps, also to federal programs that were providing food or formula to these folks.

    Has the committee made any recommendations

with respect to that, immunosuppressed populations here within the United States?

    DR. BUCHANAN:  Working in conjunction with CDC, when we start to have outbreaks associated with this organism, we have already communicated some public-health messages out during and with that process.  As Jeanne indicated, we are looking further into additional methods and different vehicles for those messages currently.

    We have also looked at the international front in terms of our activities associated with the International Code.  So we are looking at it at multiple functions.  We did, immediately, working with CDC, as soon as we started having outbreaks, get public-health messages out.  We always refine those over time.

    DR. WASLIEN:  This is Carol Waslien from the University of Hawaii.  This brings to mind some of the things the U.S. ran into with problems of Vitamin A in infant formula that the U.S. was distributing through its AID Program.  Even though this is a question for consideration inside the

U.S., the fact that U.S. is essentially distributing a product that might be dangerous out for other countries puts it into the framework of something that the U.S. should consider even though it might be that the target audience is not U.S.

    DR. MILLER:  Would it be fair to say that there is a general consensus on the committee that CFSAN ought to pay increased attention to making sure that educational materials are provided on the international.

    DR. WASLIEN:  Yes.

    Any other questions before we move on?

    Thank you.

    Dr. William MacLean of Ohio State University has a public comment.

    Dr. MacLean?

Public Comment

    DR. MacLEAN:  Thank you, Mr. Chairman.  My name is Bill MacLean.  I am here as a paid consultant on behalf of the International Formula Council which represents the formula manufacturers in the U.S.  I should also note that I recently

retired as Head of Medical and Regulatory Affairs at the Ross Products Division of Abbott and am a stockholder of Abbott Laboratories.

    On behalf of the IFC and its members, I want to thank you for this opportunity to speak to you on issues related to, as we will refer to it, the E. sak issue and  powdered infant formula.

    The subcommittee has recommended a number of intervention strategies to reduce the risk of E. sak infections, including the development of a microbiological testing program as well as the development of educational materials.  Since these recommendations were released, the infant-formula industry has provided FDA with a number of proposals and discussion documents to address these intervention strategies.

    We are also providing you today with a preliminary brochure that addresses one of the issues you have been talking about.  This is entitled Infant Feeding; Safety issues for Healthcare Professionals.  If you don't have it, it is available for you at the table outside, I


    We also understand that FDA acknowledges the interim nature of the subcommittee review and we strongly concur with the FDA's plans to continue to work in combination with industry to further address these issues.

    Today, I would like to speak about two unresolved issues, specifically the population at risk for E. sak infection and the microbiological and sampling program.  Let me take the infants at risk first.

    We agree with the subcommittee that preterm infants born at less than 36 weeks gestational age and up to  a post-term age of four to six weeks, that immunocompromised infants at any age and term infants hospitalized in Level 2 and Level 3 neonatal intensive-care units are at risk and that this risk is low but unquantified.

    As the subcommittee suggests, we do not believe that this low but unquantified risk extends to healthy term infants.  The presence of extremely low levels of the organism as reported by the FDA,

the group in Canada, the group in Holland, does not begin to suggest that an infant formula used as recommended is unsafe for term infants.

    There is a near absence of reported cases of E. sak-related illness despite feedings to millions of full-term infants each year.  The only reported occurrence of disease in healthy term infants in which reconstituted formula has been contaminated also had evidence of external contamination or mishandling of the reconstituted product.

    Furthermore, these infections are almost exclusively associated with preparation and use of reconstituted powder formula in hospital.  Thus, FDA should not extrapolate to term infants any assessment of risk derived from disease among immunocompromised and low-birth-weight infants.

    The subcommittee recommended, and we agree, that "Every effort should be made to avoid feeding powdered infant formula to these at-risk infants."  However, as Dr. Thureen commented during the subcommittee's deliberation, it is important to

maintain perspective on the low risk of E. sak infection relative to all the other more common pathogens and other hazards to which preterm infants are exposed in the neonatal intensive-care unit.

    We would like to reemphasize that, in many cases, there is a medical need for powdered products.  Powdered products offer a significant benefit to the 10 to 15 percent of babies who are born prematurely and/or with low birth weight.  Powered human milk fortifier is a prime example of this.

    Human milk is known to offer significant advantages to preterm infants yet it is nutritionally inadequate for preterm infants unless it is supplemented.  The availability of these and other powders offer neonatologists flexibility to provide individual nutrition to these low-birth-weight infants that simultaneously accommodates things such as fluid-restricted diets, delivery of specific nutrients and desired improvements in growth.

    These medical benefits must be weighed against the very low risk of E. sak-related disease in the at-risk population.  Any decision to remove powder from the NICU would place neonatologists in an untenable position.

    Risk management should focus on formula handling, especially after reconstitution, not the elimination of powdered infant-formula.  Despite continuous improvements in manufacturing, it is important to recognize that manufacturing changes are going to play a relatively limited role in risk management.

    FDA also charged its surrogate to identify intervention strategies that can be used in

infant-formula manufacturing processes and plants.  The subcommittee encouraged the development of a microbiological sampling and testing program through joint efforts by industry and FDA.  The infant-formula industry has provided FDA with two possible options to assess microbiological control.

    A workshop was held on November 7 of this year in which some differences in laboratory

procedures used by FDA and industry were identified.  Alignment of testing methodology in use by both FDA and industry is essential.  Pooled industry data revealed about a ten-fold lower prevalence rate of E. sak than FDA's 2002 field survey.

    There is no validated method to test for E. sak in powdered infant formula.  The current FDA methodology detection limits are inconsistent with its zero-tolerance policy.  Newer methodology, for example, for conformation of isolates using DNA showed that 39 percent of 77 samples identified using conventional testing methods were, in fact, not E. sak.

    Thus FDA and industry's employing a common validated methodology will ensure that E. sak in powdered infant formula is consistently and accurately monitored and reported.

    It is essential that we continue to further define and reduce, to the extent possible, any potential risk posed from powdered infant formula while assuring that this product, which is

essential for the health of infants remains available.  Efforts by industry, academics, practitioners and regulators should continue to include actively collecting information about the organism, itself, including all the possible ways infants might be exposed to the microorganism and the frequency and incidence of its occurrence.  A sound risk assessment can only be conducted once the infectious dose is known.

    Mr. Chairman, if I might, it was not part of my original remarks but there has been a lot of discussion and comments about the issue of powdered formula in Africa.  If I might, I would offer an answer, a personal answer, as a pediatrician who has worked in malnutrition outside the U.S.

    DR. MILLER:  If you can do it two minutes.

    DR. MacLEAN:  I can do it in two minutes.  Dr. Dwyer raised the issue--my understanding, Dr. Dwyer, is that most of the powdered formula that is used in AIDS in Africa, for example, is infants who have been born to HIV-positive mothers, most of whom, or many of whom, have been dosed with drugs

at or around the time of delivery and, therefore, the goal is to prevent vertical transaction to these infants through breast feeding. In other words, these babies are, in fact, generally healthy term infants who do not yet have HIV.

    The second point I would make is that the risk of using any formula but, in particular, powdered formula, be it in these infants or normal infants in developing countries, the risk of contamination through water and reconstitution is far, far greater than the risk of E. sak.  Just to put that in perspective for you, the detection of E. sak in powdered infant formula in this country is about 0.36 colony-forming units per 100 grams which means that, in two-and-a-half to three days of feeding, an infant would get one organism which is many log orders below what they would need based on what we know for infection rates.

    The key thing is, after it is reconstituted, to feed it before the organism can go into a growth phase.  If that is done, my view is that the risk would be relatively low.  The

issue for formula for any of the pathogens is dirty water, improper reconstitution and then storage without refrigeration.

    Thank you, sir.

    DR. MILLER:  Thank you.

    Are there any questions for Dr. MacLean?

    DR. DURST:  Durst.  Can you provide the reference on this new methodology for confirmation of the isolates?  It is 16SR RNA.  This is not always the most specific methodology and I was just curious is there is a reference available on that.

    DR. MacLEAN:  I believe there is.  We will certainly provide you what we have.

    DR. DURST:  Thank you.

    DR. MacLEAN:  I am not a microbiologist so I cannot answer your question beyond that.

    DR. MILLER:  Dr. Dwyer?

    DR. DWYER:  Thank you for those clarifications, Bill.  I certainly think you are absolutely right about getting validated and agreed-upon methods.

    The other issue that you spoke to and we

got a chance to at least glance at the brochure is how important it is to have education of everyone in the NICU, at least in this country, where many of those babies are residing.  I think it is very useful that you went with the dieticians who are especially concerned about food and see that as sort of their function.

    It is also important, of course, to reach the NICU nurses and the neonatologists.  It seems to me that the detail people who go around to these places all the time anyway from the different companies are particularly good a delivering messages like this.  Maybe that is a useful partnership for the public health.

    DR. MacLEAN:  Yes.  I agree with you entirely.

    DR. MILLER:  Other comments or questions?  If not, thank you.

    DR. MacLEAN:  Thank you.

    DR. MILLER:  That completes our discussion of the report of the Contaminants and Natural Toxicants Committee.

    Before we move on, I understand that Mr. Levitt is here and is going to make his delayed opening remarks.  Just make sure you put it in the right place in your notes.

More Opening Remarks

    MR. LEVITT:  Thank you very much.  It is a pleasure to be here and to see so many of you.  It was just a couple of weeks ago where we were very pleased to send around, and I hope people got copies, our end-of-year accomplishments for CFSAN for 2003 where, again, we were pleased that we not only reached but exceeded our goal of completing over 90 percent of our priorities for the year.

    That is not only the second year in a row we were able to do that, but, as we went back and looked at the actual number, not percentage, we actually mapped out that, over the last five years, we have actually gotten more done each year and we feel very good about that.

    But I gather that news got a little bit foreshadowed a week later when I announced that I had decided to retire from the federal government

and go back to my roots and be a lawyer.  Hard to imagine, by some, I guess.  But I will be retiring at the end of this month and begin a new career in January.

    So I am particularly pleased, as I try to go around and reach closure in different forums, that Food Advisory Committee was meeting this month because it gives me an opportunity to directly thank you, say good bye and give you a few thoughts of my own.

    Over the last six years, one of the things we really set out to accomplish was to strengthen the scientific base underpinning and stature of the Center for Food Safety and Applied Nutrition.  I remember, just after I got appointed, one of the first people I spoke to was Sandy Miller.  Sandy was frank, as always, and said, "Joe; it's the science.  It's the science."  Right, Sandy?  That's what he said.

    DR. MILLER:  You were a lawyer, you see.

    MR. LEVITT:  That's right.  You see, it all kind of comes in a nice circle.  We have, I

think, made very, very good progress in that area.  We started this fiscal year about 200 people above the number we had when I started.  When I came, it was about 750 people in CFSAN.  We started this fall, this fiscal year, at about 960.

    That is the net increase.  The actual number of new people, because a lot of people left who retired, it is more like 300, 350.  We have been able to recruit a wide cadre of people with strong scientific expertise, a high number right out of college or out of graduate school, a number of others with much more experience.  We, I feel, have infused a strong level of science throughout the organization.

    Dr. David Acheson, who you will be seeing this afternoon, is certainly one of those and I think is an excellent example of the kind cadre of people we have been able to attract to CFSAN.  I feel good about that.  Together with the scientific stature goes with the importance of follow-through and productivity. As I said, with our accomplishments, I feel we have been able to do


    Part of all this, of course, is the work of the Food Advisory Committee.  I believe that the work this committee does is critically important.  It helps in adding openness and transparency to the work we do.  It certainly strengthens even further the scientific expertise that we bring to our decision-making and it provides an additional level of independent advice, which we need and we value.

    We have also, as you know, greatly expanded the Food Advisory Committee.  When I came, there was one standing advisory committee.  I remember my first meeting was on Olestra, so we got a good initiation on that but really saw that there was a value in greatly expanding the reach of the advisory committee and went out and have established six subcommittees a number of which a number of you chair.  I want to thank all of you who are already giving reports on today.  But we have subcommittees on Infant Formula, Contaminants and Natural Toxicants, Food Biotechnology, Food Additives and Dietary Supplements and a new one

which we have just established and will be meeting this year, we hope, in Nutrition.

    So, again, I thank everybody for your work on there.  I really believe that the role of this kind of committee and the subcommittees and the public discussion of the very important issues we face is an enormous benefit to the FDA and to the public.

    We have, just within the last year, taken on, or a couple of years, a number of very important issues, as you all know, and have participated in.  Methylmercury, which was probably one of the, I think, most visible in watershed meetings on the Food Advisory Committee and I think it is appropriate that the last meeting I am able to address is also revisiting methylmercury.

    I will those this afternoon go into the details of it but we did very much take your recommendations to heart and have gone back and tried to answer those questions in terms of what does it mean to eat a variety of foods and what is the contribution and the role that tuna, in

particular, plays and what are the scientific exposure and risk assessments, what is the best advice to children.

    I think you will see we have come back and done our best to try to answer those questions, work them into advice and probably, maybe even most importantly, work hand-in-hand with EPA.  I remember that was one of the significant recommendations that came out and I think you will see FDA and EPA standing up here together.

    As before, we will continue to welcome additional advice you have.  It is hard to get an issue of that level of complexity exactly right.  I have no doubt that people will have additional thoughts as we try and provide the very best advice to women that we possibly can.

    We have also taken on other issues; acrylamide, E. sakazakii, that I just heard some discussion about, latex, food biotechnology, dietary supplements.  These are all issues that are ongoing and we will need your continued advice and counsel.  And I thank you for that.

    I want to give particular thanks to our distinguished Chair, Dr. Sandy Miller.  Sandy has provided, for me, personally as well the Center, a number a important roles.  When I first came to FDA, Sandy, was, of course, a Director of the Center for Food Safety and Applied Nutrition, or I guess it was the Bureau of Foods, at the time and actually renamed it Center for Food Safety and Applied Nutrition.

    So, when I first came to FDA, that was my model of what the Food Center ought to be and what the Director ought to be.  While I doubt I have been able to measure up entirely, I have tried to at least follow with the degree of trying to engage on the issues that I know Sandy always brought when he was at FDA, and has ever since, and, as Chair of this committee, you have brought just the right level of, again, continued candor, focus, scientific expertise and discussion of the issues in an open form and I thank you so very much, as I know does the entire Center.

    DR. MILLER:  Thank you, Joe.

    MR. LEVITT:  I want to extend thanks to the different subcommittee chairs.  You may not have known what you quite signed up for but, again, I think you will find the time invigorating and rewarding and probably a little frustrating, too.  That is okay.  That is part of the picture.

    One thing you are also finding at FDA is, while I, as an individual, am retiring it is also a time at FDA where there is a significant turnover for a number of different administrative reasons.  Actually, with CFSAN, there are about 50 people that are retiring at the end of this month or during this month.

    One of those people I want to recognize here that has been very close to the members on the Food Advisory Committee is Cathy DeRoever.  Cathy has been serving as the Exec Sec kind of since time immemorial, it seems.  But, as you know, there isn't anything of this magnitude that happens simply.  Putting together an advisory committee is a tough job.

    You have got to, first of all, find the

right people and do all the right background checks and be sure that we have the right expertise and the right balance and the right focus from lots of different quarters.  You have got to put together these meetings which are no mean feat.  Just getting a date for the meetings, itself, deserves a Nobel Prize, I think.  But to try to get the facilities and the hotels and the accommodations and the schedules for everyone is no small feat.  And Cathy has done it with extraordinary capability and with grace and I want to publicly thank you, Cathy, for your contributions here and in many other areas throughout your career at CFSAN.  So I want to extend our personal thanks to you, also.

    Finally, I will want to do here what I am trying to do in as many fora as I can which is to help what I hope will be a very smooth transition at the helm of CFSAN.  When the Commissioner, Mark McClellan, came out the day that we announced my retirement to CFSAN, had four main themes that he tried to enunciate and which I will try to reiterate and try reinforce.

    The first is that the Food Center is doing well, that we have amassed a strong record of accomplishment and he feels very good about that.  As you know, I feel very good about it, as well, as I think do many of our stakeholders and staff members.  So it feels good for me kind of going out when the program is at least, I can't say on top because I expect it to be even butter, but certainly doing well.

    Second, and, again, this is a good feeling for me but I think an important message for everyone that his charge to the Center was continuity, continuity of priorities, continuity of the way we have been conducting business.  The FDA announced this past summer a Strategic Action Plan which goes to a number of cross-cutting areas and the Food Center is heavily involved in virtually every single one of those.  So I think the theme of continuity is also important.

    Third is smooth transition.  I am working with Dr. Bob Brackett very closely--he is here.  I am going to announce and introduce him in a minute

and give him a chance to stand up and say hello and introduce himself--ut working very hard within and without so that continuity a priority, to have that continuity of leadership.  I think people know that I was very involved with bringing Bob to CFSAN.  So, for us, that smooth transition feels very comfortable.

    But I want to be sure we are reinforcing it.  In keeping with that, finally, I think the future--and I know the Commissioner shares

this--the future of CFSAN and foods  programs at FDA is a bright one.  We have put in place a lot of things built on strong traditions in the past, strength in the science and sheer productivity, bringing in outside advisors.  I think I look forward, in my new capacity, to both viewing that and contributing in different ways.

    With that, I just want to take a moment and introduce and invite up to the podium so you can meet him directly Dr. Robert Brackett who will be the new CFSAN Director starting January 1.  Bob is already here but he is going to have to wait and

let me just introduce him for a second.

    Bob came to CFSAN in the spring of 2000.  Before that, he was actually a member of the Food Advisory Committee and those who are on the committee, then, for a short time.  He didn't last long because we gobbled him up.  Sometimes you see somebody and you say, this guy is too good.  We can't let him get away.  We knew that he was one of them.

    Bob has a Ph.D. in microbiology from the University of Wisconsin.  He came most recently from the University of Georgia, Center for Food Safety, under Michael Doyle's operation there.  I am sure many of you know him.  He has co-chaired, over the last couple of years, in his capacity as Director of Food Safety in our Center, the Joint Advisory Committee that we have with USDA, what is called the National Advisory Committee for Microbiological Criteria for Foods.  Somehow, somebody came up with the acronym NACMIF.  I first said, what is a NACMIF?

    But it is a strong joint committee and Bob

brings his expertise in co-chairing it.  Within the Center, he began as a senior microbiologist within our Office of Plant and Dairy Foods and Beverages, what we finally referred to as Landfood.

    He quickly moved up to be Director of Food Safety.  Last spring, as the issues of counterterrorism rose to the fore, we added to his responsibilities to make him also Director for Food Security as well as Food Safety.

    In my dealings with him, Bob has, I think, simply exemplified the very best of what we have tried to recruit into CFSAN and bring strong science, strong integrity, clarity of thinking and, above all, a desire to do what is good for the public good.

    So, while everybody knows no one is allowed to choose your successor, but you are allowed to take a great pride in one when someone such as Bob is named.  So, please, give a warm welcome to Dr. Bob Bracket.


    DR. BRACKETT:  Thank you, Joe.  I would

like to say hello to all of you.  Many of you I know quite well.  Many of you I don't know yet but I do look forward to getting to know you better on this committee.

    As Joe said, I was a member of the advisory committee about the time that they tried recruiting me.  In fact, Cathy DeRoever and I spent time in the hallways between some of the meetings talking about things that we might do.  So I am very, very supportive of the advisory committees.  I think the scientific advisory committees are to our policies and to our science what the journal management and editorial boards are to scientific journals and I think that is very important, as Joe mentioned.

    So I will say that I will look forward to meeting all of you.  Please feel free to call on my if there are things that I can do in my new capacity.  Joe has left a very big set of shoes to fill but, at the same time, he has done everything he can to make it as smooth for me as he possibly could and I do thank him for that.

    DR. MILLER:  Thank you, Bob.  Thank you, Joe.  The torch passes on.

    I was going to ask to see if anybody had any questions for the new Director, but I decided discretion was the better part of valor in this case.  Let's keep everybody friendly for the moment.

    The next report is on Dietary Supplements.  Johanna Dwyer is going to make a presentation.  What I normally do is try to ask somebody to make a brief note of the discussion.  I have asked Daryl Lund to make the one on Contaminants and Natural Toxicants.  But I think it might be useful if we ask the Chairman of the subcommittee to do that for us.  So, Johanna?

Dietary Supplements

    DR. DWYER:  This committee had a very interesting challenge.  We met in March.  If you look in Tab 1, you will see all of the details of the meeting.  But the basic question had to do about the definition of metabolite.

    The first question we were asked to answer

was whether it was possible to identify a particular set of scientific criteria or principles to determine whether a substance was or was not a metabolite of another dietary ingredient.

    Dr. Dickinson was there.  I had a wonderful committee to think about this with but I have to tell you, Dr. Miller, all through this meeting, I kept thinking about you from many, many years ago when you were a Professor at MIT and I was privileged, or whatever, to take your course.  You always asked us questions like this; what is a metabolite.

    Thanks to several members of the committee, we did come up with a definition.  It is a wonderful question for those of you who have to do oral exams for Doctoral students.  We decided that X is a metabolite of Y if ingestion of Y by humans results in net production of or increased flux of X incorporating structural elements of Y.  Now, if that doesn't sound like E=mc2, I don't know what does.

    But if you think about it and reason about

it, you will see that it pretty much does cover the questions that we were asked to answer.  We do have a few caveats there that we have listed in depth in the report.

    The second question we were asked

to--there is one interesting one that I call to your attention and that is whether human ingestion included or did not include changes in substances by the microflora of the gut.  We decided we thought it did.

    The second question we were asked, again focusing on what is a metabolite, was whether the strengths and weaknesses of the concepts, whether there were some concepts that were necessary to consider with respect to their usefulness in identifying whether a substance was or was not a metabolite of another dietary ingredient.

    One criterion was direct or indirect participation in catabolic or anabolic sequences or pathways.  The second issue we were asked to consider was proximity.  The third, in terms of semblance to another dietary ingredient with

respect to structure, function or a combination of them and, finally, qualities or similarities to another dietary ingredient relative to speed or rates of different sorts, compartmentalization or fate.

    Going back to that definition, we agreed that the items that were just mentioned did seem to be adequately addressed by our definition.  We didn't agree as to whether metabolites, at least in common parlance in the science that most of us were familiar with, were both catabolic and anabolic.  There was some question about whether you called things that were products of anabolism and metabolites.  But basically we did come to agreement on that, too.

    Then the third and final question we were asked was about the scientific validity and likely utility of the concepts that we had discussed when identifying when a substance wasn't a metabolite or was of another dietary ingredient and what characteristics associated with the criterion made it valid or useful.

    We agreed that the definition we sort of forged emphasized product precursor relationships, metabolite formation as a result of human ingestion, net production of the product from the precursor and also it incorporated structural elements.  So we were quite pleased with our definition.

    I would like to recognize Dr. Bonnette who was our officer on this and who helped us enormously and see if he wants to add anything to the report or if Dr. Dickinson who was present at the meeting does.

    DR. DICKINSON:  I do not.  I think that is very complete.  Thank you.  That was Dickinson.

    DR. MILLER:  Does that complete your report?

    DR. DWYER:  That completes it.

Questions of Clarification

    DR. RUSSELL:  Johanna, this is Dr. Russell.  I noticed there was this criterion that had been requested about the safety component and then several members, it says at the last

paragraph, repeated reservations about lack of a safety component and the definition.  Can you expand on that a little bit more, what the arguments revolved around?

    DR. DWYER:  Yes.  I defer to Rich Bonnette who is very well familiar with this.  I was one of the persons who raised those concerns, as I remember from the transcript.  I believe the issue there had to do with whether--it really was a legal question more than anything else.  As I understood it, the answer we were given by the experts who were there was that other parts of the FD&C Act and other regulations would cover the safety issue and, therefore, we didn't need to be as concerned about it for this specific thing.  Dr. Bonnette, do you want to add anything?

    You don't look like Dr. Bonnette.

    MS. LATHAM:  No.  This is Jeanne Latham.  I am actually now the new Dietary Supplement Executive Secretary.  The Executive Secretary who was there was Constance Hardy.

    DR. DWYER:  Oh; I'm sorry.  I'm all mixed


    MS. LATHAM:  You are exactly correct.  The discussion points at the meeting included the fact that the safety is part of another section of the Act and we were specifically focused on the definition of what is a dietary supplement or a dietary ingredient.

    DR. MILLER:  Dr. Callery?

    DR. CALLERY:  Callery.  I wonder if we could get a little more background as to maybe beyond what was just said about why the question was asked of what is a metabolite.  I have a follow-up to that.

    DR. MILLER:  Annette?

    DR. DICKINSON:  The reason the question comes up is that, under the Food, Drug and Cosmetic Act, the definition of a dietary supplement includes a specification of what kinds of ingredients it might contain.  These include vitamins, minerals, botanicals, various other ingredients or metabolites of those or concentrates and other kinds of things.

    So the question for the committee was how broad is that.  What does that mean, or metabolites of those?  So, for purposes of qualifying as a dietary supplement, the issue was what is the scope of term metabolite.

    A second question which Johanna mentioned and which is mentioned in the minutes, was we also have other concerns about those ingredients; for example, are they safe, are they beneficial, and so forth.  The nature of this discussion, however, was to answer the definition question and there are other sections of the Act that relate to--even if you accept that it meets the definition of an ingredient, there are still other criteria about whether it is safe to use and whether it may be used.

    DR. CALLERY:  Then let me add this comment to it.  I come from a drug-metabolism background.  Much of this definition seems to follow that type of thinking, what happens to the substance once ingested in the body, or what does the body do to the substance.

    I have colleagues in phytochemistry and botany and drug-discovery people who think of secondary metabolites in plants as a research for biologically active substances so that the whole definition of metabolite to them is quite different from the human conversion type of thinking.

    I will give you an example.  Out of opium are metabolites--well, morphine is one of them but codeine is another metabolite.  Here is an indication that, in fact, if you said that that was only one substance, then you would not have the definition that they are looking for.

    So I am wondering if the definition wants to be expanded to other metabolites of known ingredients that you can still follow the same definition that it is enzymatically formed, et cetera, but it may be the enzyme that was in a plant, for example, rather than in the human.

    DR. DICKINSON:  Right.  There was discussion of the fact that we had to take into account that there are also plant metabolites of components in those plants.  But this had to do

with an actual ingredient that is added to a supplement.  First of all, the ingredient that is added must meet one of the criterion in the definition; that is, it must be a vitamin, mineral, botanical, et cetera.

    The further question then would be what kind of metabolite of that ingredient could also be added as an ingredient of a supplement.  In other words, the metabolite you are talking about would be a metabolite already produced by the plant.

    DR. CALLERY:  Right.

    DR. DICKINSON:  And, therefore, theoretically, would already be eligible for use before you take into account the other restrictions on what can be used.  The question here is if you took a substance like that and, perhaps, in a laboratory created another metabolite of it, what would the scope of that term metabolite be?  In other words, the ingredient that is the metabolite of the plant is already an eligible ingredient.

    DR. CALLERY:  As a botanical.

    DR. DICKINSON:  As a botanical.

    DR. CALLERY:  Maybe you want to call it human metabolite.

    DR. DICKINSON:  We did specify that it is a human--that metabolite for--

    DR. CALLERY:  That metabolite means human metabolite.  Then another is the toxicology part is the conversion of metabolites that make toxic products where the precursor is not toxic.  Is that now part of your definition of metabolite for the conversion to a reactive species that may not be necessarily identified?

    DR. DICKINSON:  And that might be toxic?

    DR. CALLERY:  Like you have a metabolic pathway.

    DR. DWYER:  I don't think we got into that, did we, Annette?  I believe that we didn't specify whether something was "good" or "bad," but rather the process.

    DR. DICKINSON:  That's exactly why we got into the safety question because, obviously, there would be metabolites of some components that would be clearly unsafe.  The charge to the committee was

not necessarily to draw those distinctions but to, in the first place, say what is a metabolite and then it would be subject to other provisions regarding safety as to whether it could actually be used.  In other word, its safety or non-safety did not affect whether it was a metabolite.

    DR. MILLER:  Do you think it would be worthwhile to recommend to the subcommittee to review the definition and clarify it?

    DR. CALLERY:  I think that it would cause confusion to the plant metabolite people if you don't say that it human metabolite and maybe just adding metabolite by making it clear up front or always that it is a human metabolite that you are referring to.

    DR. DICKINSON:  I think that is included in the definition where it says, if ingestion by humans results in--

    DR. DWYER:  If the Chairman would allow us--if I could get your ideas on the definition, we can certainly put that in the report.

    DR. MILLER:  Okay.  Other comments?

    DR. DURST:  Durst.  Just a curiosity on the definition why it is necessary to have net production and increased flux of X.  What is the significance of that distinction?

    DR. DICKINSON:  We had a guy there who knew that but it is not necessarily one of us.

    DR. DWYER:  I can tell you, your questions are very perceptive.  One of the leading lights in terms of thinking this through and arguing it through was Dr. Eric Brass who is a pharmacologist.  I believe that the transcript, the complete transcript, will get at those points and which part of the discussion led us to specify both increased production and the increased flux.

    DR. MILLER:  Dr. Waslien?

    DR. WASLIEN:  One of the questions is efficacy; correct--because then you would say it had to have an increased production, if you were looking at efficacy.  But if all you are looking at is a source, you wouldn't have to have that flux, that increased production.  Waslien.

    DR. CALLERY:  Callery again.  Since we got

involved in editing a little bit here, I would like to add another need for clarification when you ask, does not violate principles of stereospecificity makes little sense.  I would recommend that that one be deleted.

    DR. DWYER:  Just share your deletions and emendations with me and I will put it into the report.

    DR. MILLER:  These are all useful recommendations to go back to the committee.  I assume that this would have to go back to the committee for their agreement.  Any other comments or questions?

    DR. CALLERY:  I am not sure we have to hold it up.  If that holds it up to do these, which I think are minor changes--

    DR. MILLER:  I am not prepared to tell the staff how to handle this, but one way of doing this is by mail.  The committee signed off on a particular report and we are making changes and they don't seem to do violent damage to what they were trying to do.  It is a question of

clarification.  I think it can be easily done by mail.

    That, I think, is one of the functions of this committee.  Our role is, I think, not to begin debating the science, although that is what the subcommittee did.  There is no point in repeating it when it is just kind to oversee this to make sure that their results are clear and are doable and useful.

    If there are no other questions, we are going to move on to the next presentation which is Additives and Ingredients.  Dr. Johanna Dwyer, again.  You will take some notes on this?  Again, I am going to ask to inform you all that please identify yourself for the record before speaking.

Additive and Ingredients

    DR. DWYER:  Thank you.  This is also under Tab 1.  This is Dr. Bonnette.  I finally got it right.  This was in August and it was on

food-mediated latex allergy which is a very interesting issue.

    The questions we were asked in this

meeting were, first of all, has a positive relationship been demonstrated between the use of natural rubber latex gloves in food service and allergenic reactions to food served in food establishments--in other words, in the eaters, not in the people that are wearing the gloves--or sold at the market based on available data.  Then, if it did exist, what was the strength of the relationship and was the relationship causative.

    We heard a lot of testimony and listened to a number of different people.  We concluded and agreed that the scientific evidence demonstrates that natural rubber latex allergens, and there are several of them, can be transferred to food from natural rubber latex gloves.

    However, we also agreed that evidence was very weak whether these allergens were transferred to foods in amounts that were sufficient to elicit systematic allergic reactions in variably sensitive individuals.

    We also felt that the absence of any good information about dose--we didn't hear very much

about dose--that was required to elicit allergic reactions in sensitized people made it awfully difficult to extrapolate whether such a transfer would have any clinical significance.  So, yes; we thought that it could be transferred but we weren't sure that it was clinically meaningful.

    So we suggested a weak positive relationship between the use of these natural rubber latex gloves and food-mediated latex allergic reactions.  We felt that the data linking the presence of these proteins, these allergens in food to allergic reactions, was based primarily on anecdotal evidence and was weak.

    That brings us to the second question which was whether a positive relationship had been established.  If it had and been shown to be causative, then could we suggest some science-based options to mitigate food-mediated latex allergy risk.

    As I have already said, we really felt that the evidence was sufficiently weak to preclude us recommending a ban on latex gloves.  In other

words, several people suggested that we do that but we didn't think that was warranted because of the level of evidence being weak.

    However, we did agree that there existed other evidence suggesting that the cornstarch, which is called donning powder--if you use these gloves, you know that a lot of times, you use cornstarch--serving as the vehicle for the transmission of allergenic proteins.  So the donning powder gets a lot of the proteins in it.

    Then, if you could eliminate that powder, it would eliminate a lot of the exposure to the allergen.  So we suggested adoption of a

food-service glove standard that specified glove powder or protein limits on food-service gloves.  We also acknowledge, though, that the data to properly establish such limits really didn't exist.

    We thought, also, that this kind of standard should apply to all producers of food and it shouldn't be limited to retail food outlets and food service.  In other words, if this is true, what's sauce for the goose is sauce for the gander

and it also applies in processing for frozen foods or something like that.

    We suggested that, perhaps, there was a need for a public education-project to heighten public awareness of this problem.

    Our third question was if the evidence wasn't sufficient to establish a relationship, what other kinds of questions did we need to ask to understand the issue.  What we felt very strongly about was that a well-designed, double-blind, clinical, placebo-controlled, low-dose,

oral-challenge study would be necessary to really find out the threshold doses for the kinds of reactions that people were talking about and to better understand what percentage of the latex allergic population was really at risk of a reaction to food contaminated with the gloves.

    So we suggested that that be done as well.

    I think that is about all I have got to say on that.  Does anyone have any other suggestions or comments?

Questions of Clarification

    DR. LUND:  Daryl Lund.  Johanna, is the only allergic reaction to the latex gloves caused by the donning powder?

    DR. DWYER:  No; I don't believe so.  I think it is--

    DR. LUND:  There are other allergenic constituents in the latex gloves, themselves, that cause the allergic reaction.

    DR. DWYER:  Yes; it is just that it concentrates in the donning powder.

    DR. KRINSKY:  Norman Krinsky.  I am a little bit confused, Johanna, because it seems to me that your answer to the first question indicates that there is not adequate evidence.  But your answer to the second question leads to making recommendations.  That doesn't seem to make sense to me.

    DR. DWYER:  Thank you, Dr. Krinsky.  We felt that there was a positive relationship but that it was weak,  that the positive relationship was weak but that there was a remedy to making it even weaker.  Dr. Bonnette, do you want to speak

with knowledge of the transcript?

    DR. HEPP:  If I may, my name is Mark Hepp.  I participated in this meeting.  I think that most of the committee felt that allergic reactions elicited by people consuming food that was handled by latex gloves were plausible and that they also recognized that the data indicated that the main part of the exposure would result from the

natural-rubber proteins combining with the donning powder and being transferred to the food.

    There was data presented at the meeting that showed that powder-free gloves transferred undetectable levels of protein.  So, the committee, feeling that the relationship was weak but because some protein could be transferred to food made it somewhat plausible, they felt that, by recommending specifications on glove powder and protein levels that, although there wasn't sufficient data to take regulatory action, they could address a potential problem or something that seemed plausible, anyway.

    DR. DWYER:  I think it is also true, and, again, my memory may fail me, that the gloves vary

widely.  There is some reason to think that, by source, by the manufacturer or the country of manufacturer, there are quite big differences in the amount of allergens present.  Is that correct?

    DR.  HEPP:  That is correct.

    DR. DWYER:  So we felt a standard that tried to get it down across the board would be helpful.

    DR. MILLER:  Dr. Krinsky, do you want to continue?

    DR. KRINSKY:  Norman Krinsky.  I just wonder how industry has responded, then, to this recommendation.

    DR. DWYER:  I am unaware of the response so I can't answer that, Dr. Krinsky.

    DR. MILLER:  Jean?

    MS. HALLORAN:  Jean Halloran.  It is my understanding in the case of peanut allergies that the threshold dose varies enormously from one allergic individual to another.  Some people are extraordinarily sensitive, others not so sensitive, by orders of magnitude.  I was wondering if you had

any discussion of that in terms of designing the oral-challenge study and getting a big enough sample or is there literature so that you could select a range of individuals or is there information on how to--the percentage of individuals who are highly sensitive versus less sensitive?  Or did you not get into that?  Would that be part of the well designing?

    DR. DWYER:  Those are very good questions.  The thing is that the data, it didn't seem to me and I speak as someone who is more familiar with peanut allergy, that the databases were as well developed as they were for peanut allergy.

    We did hear from a number of very interesting--much interesting testimony.  We did hear from a few physicians who were dealing with a highly allergic population and who one would presume would be the people who were the most sensitive.  Even there, the dose-response studies were not all that the committee felt were good ones.  They were not as well developed.

    So the answer is there are a lot of

unanswered questions here but the relationship didn't seem, after we looked at everything including a great deal of testimony from people who were among these individuals who were presumably highly sensitive, we felt that the relationship, if there was a weak one.

    DR. MILLER:  Dr. Downer?

    DR. DOWNER:  Goulda Downer.  I participated in that meeting, also.  I think one of the things that also clouded the discussion was that individuals who are allergic to natural rubber latex also had an allergy to kiwi fruit and bananas.  So the challenge was how do we know exactly what they would be allergic to.  We recognized then that they did have some allergy to the latex but that was something that made it even more confounding.

    We didn't have enough data.  We did have some individuals who were very persuasive in explaining some of the challenges they had or patients had, becoming very ill.  However, there were some other confounding variables clouding


    DR. MILLER:  You mean that they were cross-sensitive to kiwi and bananas?

    DR. DOWNER:  Yes; they all were.

    DR. MILLER:  That is interesting, in itself.

    DR. DOWNER:  Yes; it was.

    DR. DWYER:  I think they are the same family, aren't they, botanically.

    DR. MILLER:  Other comments or questions?  Yes?

    DR. SCHERER:  Cliff Scherer.  I was just wondering, the committee recommended a

public-education program.  I was wondering whether any thought had been given to exactly what kind of message that might be and what the purpose of it would be.

    DR. DWYER:  We didn't really get very specific beyond the issues, as I remember, and, again, Dr. Downer or others who were there may want to elaborate, certainly, this issue of the donning powder.  To heighten public awareness, I think,

also meant heightened food-service establishment awareness.  So it was more of a suggestion to try to get people who were food-service handlers to be aware of these sorts of thing.

    DR. DOWNER:  Goulda Downer.  In addition to that, we had recommended for the manufacturers that restaurant establishments may want to voluntarily put up large signs says, we are

latex-free, so the consumer would have a choice in assisting in doing that.

    DR. MILLER:  Further comments?

    DR. WASLIEN:  I always cringe when I hear somebody adding another thing to the list of--oh; Waslien--cringe when another thing is added to the list of free environments when the scientific support for any relationship is very weak.  It would seem very, very premature to start talking about permitting manufacturers, or food establishments, to say, we are latex-free.

    I would think, and I also worry.  I am not an immunologist, but I thought I would worry about the ethics of doing exposure in individuals who are

highly sensitive to an antigen.  I think the other studies that are listed on here further on have seen the amount of latex that goes into the donning powder and how much, actually--what procedures would put it in foods--would be first-order priority before any human studies; one, because of cost, and, two, because of ethics.

    DR. DWYER:  Thank you for those comments.  I think the committee's judgment is reflected in the transcript.  We did hear testimony from, I think, one state for sure and maybe a couple where they had already put some regulations in place.  So it is critical to get a very good science-base on this before we go too much farther.

    DR. MILLER:  Would it be worthwhile for this committee to recommend to the subcommittee that, in order of priority, that the studies done on--the in vitro studies be done--they will do that, anyway, but just to emphasize the ethical issues involved.  Would that be worthwhile?

    DR. DWYER:  We will do it.  Thank you very much.  Good questions.

    DR. MILLER:  Any other?  Yes?

    DR. SCHERER:  Cliff Scherer, again.  I just wanted to make one note of thought about the problem of beginning to alert the public.  I think if we really talk about the public as different than food-service or whatever, but the idea of focusing them on what appears to be a relatively minor risk while we have others that we maybe do want them to focus on.  I think that is the tradeoff.

    DR. DWYER:  It is part of public education, too, so I am not sure our statement precluded that, that the risk might, in fact, be very low.

    But the point, again, is that there are already states that have done this.  So this is all after the horse got out of the barn, so to speak.

    DR. MILLER:  Okay.  Then why don't we move on to the last topic for the morning, Food Biotechnology.  Frank Busta, again.

Food Biotechnology

    DR. BUSTA:  Just to recognize that I will

be going fast, good afternoon.  The Food Biotechnology Subcommittee has met twice.  You have the second of the meeting reports in front of you.

    The first meeting was on August 13 and 14.  Basically, that was to discuss the science-based approaches to assessing whether new proteins or bioengineered foods are likely to cause allergic reactions in some individuals.  There was a good discussion.  It included looking at some of the Codex documents and recommending that the FDA work closely with them, with those documents, looking at some areas of allergenicity research and discussing some potential draft guidelines in allergenicity.

    The second meeting that you have the minutes on the report was in September of 2003.  I was the Acting Chair at that meeting.  We received reports on the current regulatory framework, policies and procedures going on as well as what Codex Alimentarius is doing in this regard and FDA's safety assessment process, specifically oriented toward molecular biology.

    If you note the questions looking at

molecular biology data and whether--to what extent sequencing information would contribute to the identification of newly expressed substances and especially how this information might contribute to the safety assessment.

    A lot of the subcommittee discussion was individualized.  Generally, the subcommittee agreed that sequencing the DNA insert was helpful but they really didn't comment on how this information would be used in safety assessment.  Some of the subcommittee believed that the data might be useful for safety assessment in certain situations but it was not highly specific.

    One of the subcommittee members thought that sequencing the extended regions flanking the insert site of every bioengineered food plant was useful but, again, didn't comment on how that would be phased into food safety assessment.

    Also, we were asked to look at the current approaches.  The subcommittee continued to agree that the molecular biology data specified by FDA were appropriate as part of the characterization,

that they should be consistent with the Codex Alimentarius process and they supported that approach.

    Some of the members of the committee stressed the importance of characterizing proteins expressed by the genetic-engineering process and several also recommended that the phenotype be the focus of the characterization.

    Finally, on the technicological advances, there was a great deal of discussion on the proteomics, metabolomics, approaches and that those newer techniques beyond genomics be introduced and utilized for the food safety assessment program.

    There was a letter introduced by several of the members of the committee specifically requesting information sooner, faster, and a report on--this is the last paragraph in the

report--asking for updates on what FDA was doing with the subcommittee's recommendations.  FDA indicated they would respond to the subcommittee, take that latter into consideration and respond to the subcommittee.

    DR. MILLER:  Thank you, Frank.

    Comments, discussion?

Questions of Clarification

    MS. HALLORAN:  I was wondering about your bringing attention to the fact that, in the report, various members did not comment on how the information contributed to safety assessment in regards to two sets of information.  I think, in the Codex document which I am quite familiar with, it is quite clear how they contribute to the safety assessment, that it gives you data on possible problems related to toxicity, allergenicity and unexpected effects.

    Do you think it was just sort of an oversight or a thought that everybody understood what the importance was so it wasn't necessary to discuss, or was there a--it is a curious phrase.  I am wondering why you included it.

    DR. BUSTA:  My recommendation is that, in many situations--for example, looking at the adjacent components, there was no indication of what the outcome of that assessment would mean.  If

you had those data, would it indicate that there was an additional component generated?  Would it, actually, in fact, make any difference to the outcome of what is in the specific food?

    I think that was that was the reasoning behind not knowing how it would contribute to the food-safety assessment in that some differences in that sequence may make absolutely no difference in the subsequent food product.  It may not demonstrate any change in protein.  It may not demonstrate any change in the product.  It may not demonstrate any differences as a result of a small change in the sequence.  I think that is the reason why they said they didn't see how it would fit into the food-safety assessment.

    MS. HALLORAN:  I believe the people who included this as well in the Codex document and who bring it up here do so because, indeed, in certain cases, you would not uncover anything interesting if you looked at the flanking sequences.  But there is the possibility that you might uncover the existence of a gene for toxin production or

allergen production that could, perhaps, be turned on by the promoter genes that you have inserted along with the insertion and that it is to rule out that possibility that you would want to do this piece of analysis.

    DR. BUSTA:  Again, my understanding was that those discussions were generated by the concern by some people who think the flanking sequences should be assessed.  But, as I recall, the conclusion by most of the people was that a difference there does not--it would not be unusual to see changes that have no influence at all and that it would take a great deal of additional effort without any significant yield in the

food-safety assessment.

    But that is the best of my recollection.

    DR. MILLER:  I am a little curious as to why the committee did not also continue to explore that subject.  It is quite true that, it is probably correct that, sequence changes don't necessarily mean that there are toxic responses that are going to come out, or any response at all.

    But, given the nature of the technology, it would seem to me that it would be worthwhile doing it and then using it as a basis for further study.

    DR. BUSTA:  I think that is why individuals sort of emphasized the importance of characterizing the proteins, going down the proteomic, or the metabolomic, directions rather than just the genomics sequencing.

    DR. MILLER:  The genetic sequencing, it would seem to me, would be equally important.

    DR. APOSHIAN:  It might be of interest to some of you that there are now techniques available called DIGELP, difference in gel electrophoresis.  About eight universities in the country have this kind of equipment.  I happen to know we have one, too, and that is why I am aware of it.

    Many of believe that it is not important, or we are not going to be able to tell the meaning of what happens when a gene is turned on or turned off.  What is more important is to know what the gene product is.  This particular kind of equipment

can now detect and differentiate a single protein from 25,000 other proteins.  It is completely computerized.  There is a central bank and so that, in the future, probably in a year or two--we have begun using it already for our research, but in a year or two, it is going to be available to many, many more people.

    If you are concerned with allergens or toxic proteins, as you should be, the product of the gene is more important than the gene structure, itself.

    DR. ARCHER:  Dr. Archer.  I am surprised at the outcome that I am reading for a couple of a reasons.  Number one, trying to remember back when I actually got involved in looking at these things, if, for no other reason, some of the sequence data, to me, was a precision index.  It was sort of a scorecard as to what a company had done or an individual had done in terms of trying to understand what, in fact, they had accomplished by inserting a gene or deleting a gene, what else had happened.  I used to call it a precision index.

    I guess my other question would be wouldn't a responsible individual or company know this anyway and, if that data is being generated in any event, why not put it out there.  Even in those rare instances where it might be useful, I have a hard time, basically, right now, putting all my marbles with proteomics simply because, once the protein is there, the cat is out of the bag.

    If it is a harmful protein, it is going to be there.  It is a heck of way to find out if there was a way to prevent that.  Maybe you couldn't, but, if that data is available, why not use it.

    DR. BUSTA:  As I recall, there was a great deal of discussion as to how much information various organizations submitted with their materials.  Some offered a great deal of information, lots of sequencing, and some offered a minimal amount.

    We have got a comment here from someone who really knows this.

    DR. MILLER:  All right.

    DR. JONES:  Good afternoon.  Kathleen

Jones, Center for Food Safety and Applied Nutrition.  I was involved in this particular meeting.  I think the point that Dr. Busta was trying to make was that the committee agreed that sequencing information was helpful.  It was helpful as part of the safety assessment.

    One of the individual committee members felt that not only sequencing the immediate flanking regions around the insert but sequencing enormous regions from the insertion site into the plant genome was also necessary to look for some of these translocation events.  As other members of the committee informed us, translocation events happen--they do happen in genetic engineering, but they also happen in regular plant breeding.

    They couldn't decide, or figure out, how that enormous amount of sequence data would contribute to the safety assessment.  You would generate a lot of data but it wasn't sure how much that would actually tell you about the safety of that food product.

    DR. MILLER:  The problem with that is that

there are many things that happen during normal plant breeding, the same thing that happens in genetic modification . I think this is a question of--my own feeling is that this data is generally available.  Doug, did you say that?  Somebody said that this data is available.

    It at least gives you an idea of where to look if you are doing your proteomics.  It is one thing to be able to do 20,000, or 200,000, different products but, if you had someplace to look, you would be able to focus on a particular area.

    I, personally, don't think it is going to generate a lot of useful information.  But, given the nature of debate, it would seem to me that the more data you could get, the better off you are going to be.

    DR. ACHOLONU:  This is Alex Acholonu.  My question is informational.  In this report, the FDA is requested to update the activities, biotechnology activities.  My question is is the FDA actively engaged in biotechnological


    DR. JONES:  FDA is engaged in both our own biotechnology activities, meaning the review of biotech notifications that are submitted by companies for new products derived through biotechnology, new bioengineered plants as foods.

    FDA has also been involved in the international community in working in the Codex Alimentarius Committee to generate guidances for the safety evaluation of bioengineered plants for food.

    DR. MILLER:  Does that answer your question?

    DR. ACHOLONU:  Yes.

    DR. MILLER:  Jean, did you have another question?  I saw your hand up.

    MS. HALLORAN:  I was just going to make a suggestion.  From the discussion we have had, it seems as though, perhaps, the committee report could be amended a bit because it doesn't seem quite accurate to say that there were no comments on whether the information contributed to the

safety assessment.

    DR. JONES:  There were no comments from the individual that was requesting that information.

    MS. HALLORAN:  But it seems as though there was quite a bit of discussion about whether it would be useful or not.

    DR. JONES:  There was an enormous amount of discussion

    MS. HALLORAN:  So it is a little bit sort of unreflective of the meeting to word the report in this way.  Perhaps, if you deleted those two phrases and substituted a sentence saying that there was considerable discussion and varying points of view about whether this information was useful or not.

    DR. JONES:  Certainly.  We will work on the meeting report.

    MS. HALLORAN:  Okay.

    DR. BUSTA:  Recognizing that you might get grilled, maybe that is why Mike Watson went to USDA.

    DR. MILLER:  How about that suggestion?

    DR. BUSTA:  Fine, being that the real Chair of this committee is taking notes on the comments, I am sure we can put something together and send it.  Doug is the Chair.

    DR. MILLER:  As I said, I wanted the Chairman of the committee if they would act as a kind of a rapporteur to make a brief note to report back to the committee on the nature of discussion.  So I guess it is not the Chairman.  It is going to be a member of the committee.

    DR. BUSTA:  We will do our best to get that out to the subcommittee as quickly as possible because that subcommittee was very interested in high participation.  So we will do our best to let them participate.

    DR. MILLER:  Johanna?

    DR. DWYER:  Just as a procedural issue, Dr. Miller, I would feel more comfortable as a subcommittee chair, myself, in working with FDA staff to circulate the gist of the discussion that we have had here and the very good suggestions I

think to the two committees that I was involved in.  But I feel very uncomfortable about changing the minutes or the things here.

    DR. MILLER:  No; I am not suggesting we change the minutes.  What I am suggesting is that, at the end of the meeting when it comes time to summarize the chief points that are being made, I am asking that the chairman of each to come up with a short note for this committee.  The recommendations that we make are recommendations that go back to the subcommittee because you can't change the report unless they are willing to accept the changes.

    DR. DWYER:  Thank you.

    DR. MILLER:  Jean?

    MS. HALLORAN:  One more procedural question.  There is no report from the meeting on Allergenicity?  Is one planned?

    DR. JONES:  There is a report.  The summary minutes are available on the FDA Food Advisory Committee website.  It was an oversight that that information wasn't contained in your


    DR. MILLER:  Any other comments or questions before we break for lunch?  That is one way of getting the discussion closed.  Then we will meet again at 1:45.

    [Whereupon, at 12:30 p.m., the meeting

recessed, to reconvene at 1:45 p.m. this same


A F T E R N O O N  S E S S I O N

[1:50 p.m.]

    DR. MILLER:  We will get started with the afternoon session.  There are a couple of things.  First of all, for those of you who have pagers and wireless telephones, would you please put them on vibration.  We have been hearing a few little notes.  I think it is rude to other people who are trying to listen to the discussion and I would appreciate it if you would do that.

    Secondly, let me remind you again, and we did very well this morning sticking to the schedule and I am going to try to make certain we do the same for tonight and tomorrow morning as well.  So let me remind you of that.

    For those people in the audience who have just joined us, the questioning will be by the committee.  It is a function of these meetings to provide information to the committee for it to issue its report.  We will be taking no questions from the audience.  Maybe that is the reason why they set me up with my back to the audience.

    The last point I want to make is that tomorrow morning we are scheduled to start at 8:30.  It would make life considerably easier for me if we could start at 9 o'clock.  Unless there is some objection, we will start tomorrow morning at 9 o'clock.

    Do I see any objections?  All I see is great big grins.  Thank you very much.

    This afternoon and tomorrow we will be discussing methylmercury.  This committee had the opportunity of reviewing the science, particularly the toxicology, and the derivation of the RFD for methylmercury back in November, I think it was.  The committee at that time made several recommendations, six, I think there were.

    The function of this meeting is for this committee to hear from FDA and EPA what progress they have made in response to the comments and questions that were made by this committee the last time.  In particular, the issue is the advisory concerning methylmercury in food and also one of the more important recommendations was the

suggestion by the committee that EPA and FDA work much closer together on this issue than they have.

    Based on that, we have a number of people from both agencies, as you will see, reporting on progress in this area.  The first issue is going to be the Status Report and Response to our Recommendations.  The presentation is going to be made by Dr. David Acheson of CFSAN and Denise Keehner from EPA.

Status Report to Food Advisory Committee's

Recommendations on Methylmercury in Fish and


FDA and EPA Development of Joint Advisory

    DR. ACHESON:  Thank you, Dr. Miller.  First of all, I would like to thank the advisory committee for taking the time to spend the next day and a half with us discussing methylmercury issues in relation to fish.  We certainly appreciate you doing that and we look forward to hearing your comments.

    The presentation that I am going to be giving is going to be a dual effort between myself

and Denise Keehner.  I am going to start off and then Denise is going to fill in some of the middle part and then I will come back and finish up talking specifically about our response to recommendations.

    First of all, before I get into this, I want to just emphasize three important points as we move forward in our development of our advisory.  The three things that we have really tried to do here is to keep the advisory science-based, simple and protective of public health.  That has been our goal and that is where we are striving to end up; science-based, simplicity and protective of public health.

    With that, what I am going to do is to move into the presentation and really just to iterate the purpose of the meeting which is to provide a report to the Food Advisory Committee on how FDA has responded to the committee's recommendations in developing a revised joint advisory with EPA that addresses both commercially and locally caught fish.

    The structure of the presentations over this afternoon and tomorrow--I decided I would just put this in perspective for you so you could see where we were going--right now, I am embarking on the first point, the status report of how FDA has responded to the previous recommendations.  A portion of this will be a description of the process that we have used and that is where Denise Keehner is going to speak.

    That will be followed by a review of the exposure assessment by Clark Carrington.  That will conclude today's presentations, obviously with appropriate questions.  Then tomorrow we are going to begin with a discussion of the focus-group testing that we have undertaken on the revised advisory and that will be followed by a presentation of where we have ended up

post-focus-group testing.

    So, moving into the status report, this presentation is going to, first of all, briefly cover the background of the relevant recent history in relation to the development of the current

advisory.  As I have already mentioned, we are going to go into the process that we used to get where we have ended up in relation to responding to the six primary recommendations from the 2002 committee meeting.

    I am going to go into more detail of the response to those recommendations and finally end up with the question that you already all have for the current committee.

    So, by means of background, and I will deliberately keep this brief, in 2001, FDA and EPA issued separate advisories on fish consumption.  In July 2002, the FDA Food Advisory Committee was asked to evaluate this 2001 advisory.

    To summarize what was in these two advisories, this slide summarizes what was in the FDA advisory and it simply says, avoid shark, swordfish, king mackerel and tile fish.  That part of the advisory was aimed at women of childbearing age and young children.  It also advised to eat up to 12 ounces per week of a variety of other fish.  That was primarily aimed at women of childbearing


    It recommended following EPA's advice for recreationally caught fish.  There was obviously a lot more in it but this, essentially, is the primary substance.

    The 2001 EPA advisory issued at the same time had the following advice in it; to limit consumption of freshwater fish caught by family and friends to one meal per week.  You can see that as specifics; in adults, 6 ounces cooked, 8 ounces uncooked; a child, 2 ounces cooked, 3 ounces uncooked.  This should apply to areas where states have not provided advice about untested waters.

    The advice was to check with state or local health departments for advice on waters where friends and family fish.  The target for EPA's advice was to women of childbearing age and children and the recommendation was that people follow FDA's advice for ocean and commercial fish.

    The charge to the committee in 2002, just to reiterate what that was, to put this in the context of the recommendations, the committee was

asked to evaluate whether the FDA's Consumer Public Health Advisory on methylmercury provided adequate protection for pregnant and women of childbearing age who may become pregnant.

    Based on that charge, the committee came up with six major recommendations as to what FDA should do.  I am going to go over them briefly here to put this in the context and then I will be handing over to Denise Keehner to talk about the process of how we approached it.

    The first was to better define what is meant by "eat a variety of fish."  Second, work with other federal and state agencies to bring commercial and recreational fish under the same umbrella.  Third was to publish a quantitative exposure assessment used to develop the advisory.  Fourth was to develop specific recommendations for canned tuna based on a detailed analysis of what contribution canned tuna makes to overall methylmercury levels in women.  Fifthly, to address children more comprehensively in the advisory and, finally, to increase monitoring of methylmercury to

include levels in fish and the use of human biomarkers.

    I am going to hand over now to Denise.  She will go over more of the process that we have used to address these specific recommendations.

    MS. KEEHNER:  Thank you, David.  What I would like to do is to briefly walk through some of the key process-related steps that EPA and FDA followed in getting to the point that we are at today in developing a joint fish advisory.

    In the fall of 2002, the EPA Administrator at that time was Governor Whitman and the Secretary of HHS, Tommy Thompson, exchanged letters around the issue of EPA and FDA working in a more collaborative and joint manner in developing fish consumption advisories.  The agreement that was reached at that time between these two individuals was that we would collaborate and bring commercial and recreational fish under the same umbrella advisory.

    After those letters were exchanged, there was a series of meetings.  The first was between

Dr. Levitt and Tracy Mehan.  Tracy Mehan is the Assistant Administrator at EPA for the Office of Water.  Dr. Levitt is the Director of FDA's Center for Food Safety and Applied Nutrition.

    Those two gentlemen got together and they talked about what needed to be done.  The follow up to that in February of 2003 was the formation of a joint EPA and FDA working group/leadership group that consisted of both leadership representation from the two organizations as well as scientists and staff representatives that would form basically the team that would be working on the joint advisory.

    Starting 2002 and moving into 2003, FDA, Drs. Carrington and Bolger, were involved in developing an exposure assessment for fish consumption and mercury-related exposure.  Between April of 2003 and today, we have been working very well together, the two agencies, including very, very frequent meetings.  I think we meet at least once a week and staff are meeting even more frequently than that, many conference calls, many

joint meetings, that have occurred.

    Some of the key milestones and accomplishments that we have achieved as a group, as a team, since April of 2003 including planning and completing an independent external scientific peer review of the exposure assessment and also some revisions to the draft exposure assessment based on that peer review as well as some other comments.

    We planned and we held four stakeholder meetings that brought in representatives from industry, from the consumer and health professional segment, states and tribes, to talk about where we were and to get some early input.

    We worked together to produce a draft advisory and we took it to eight focus groups in four different locations throughout the United States.  We will talk about this a little bit later.  Marjorie Davidson from FDA will discuss a little bit more and in more depth the results of the focus-group meetings.  But, importantly, we were revising the advisory in real time as we heard

what the response was of the focus groups to what our message was and what message we thought we were communicating versus what was received.

    We also planned and worked jointly together in putting together materials for this advisory committee meeting today and tomorrow.

    To briefly touch on some of the outcome of the stakeholder meetings, some of the key messages that we heard--I'm sorry.  The stakeholder meetings, we held them, and I mentioned this just a moment ago, with industry, consumer and health professionals, states and tribes.  We shared with the stakeholder group a tentative time line that included these focus-group meetings as well as a public meeting.  This was shared with an idea that we would be going out with some type of public interaction in the fall of 2003.

    Some of the key messages that we heard at the stakeholder meetings included a message about needing to recognize that science would continue to evolve and new data would continue to be generated over time and that we needed to recognize that that

new science and those new data needed to be part of planned future revisions to any type of advisory.

    But, importantly, from several sectors, there was a comment that we really did need to move forward with an updated advisory, a joint advisory, from the 2001 time frame.  There was some concern that was expressed during the stakeholder meetings about the accuracy of some of the tissue data that was used in the model, the exposure model.

    There was some concern expressed about striking the right or the appropriate balance between not discouraging fish consumption because it is an important part of the diet while still maintaining an effective message on the risks posed and the concerns about mercury in fish.

    There were comments about the time line being too ambitious and there were concerns expressed about the need to make sure that the states were really brought into what we were doing because, in the end, there is a lot that occurs implementationwise that is at the state level.

    There were several comments that were

expressed reinforcing the need for effective outreach and communication after the advisory is developed so that the message that we are communicating does get, in fact, to the people that need to receive the message and that that is an important part of achieving the public-health goal.  Just developing the advisory is not the endpoint.  It is, in fact, getting the message out.

    In September and October, we spent quite a bit of in-depth time working through--taking all the input that we had heard, taking the different perspectives within each of the agencies and coming up with several different iterations of a draft advisory.  We settled on an iteration that we thought was ready for focus-group testing and, in November, we started the focus-group testing of the draft advisory.

    The first focus group occurred in Calverton, Maryland.  As I mentioned, Marjorie will be going into more depth and giving you more information on who was there and how those were conducted.  The important message, processwise, is

that, even though we thought we had a really good advisory, when we got to the focus group, the first focus-group meeting, it was clear that some of the messages that we were trying to communicate were not getting through and we had to go back and do a pretty major revision to what we had produced for subsequent testing in the subsequent focus groups.

    What we found was that, in the subsequent focus groups, we were making some refinements around the periphery but nothing as substantive as between the first focus group and the second.

    That brings us to this meeting here today.  David has already mentioned what our objectives are in being here.  We are basically here to present what we have come up with post-focus-group testing and to hear from you as to your thoughts and reflections and on how well it responds to the comments of this advisory committee and whether we are ready to move on to the next steps in the process.

    Now, David, you are next to come up again and talk about, in summary fashion, how we have

responded to each of the six recommendations.

    DR. ACHESON:  What I am going to do is just take these through just one at a time and go into a little more detail of how we have responded to them.  Each slide, the format is the same.  I have just repeated the question at the top there in the heading.

    The first one, to better define what is meant by "eat a variety of fish."  We approached this by considering a variety of different ways to get this message over.  We considered putting in lists of fish to try to illustrate what we meant by variety and ran into some questions on that as to what to include and what to exclude.

    We then opted for more expanded language to try to explain what we meant by "variety of fish."  As Denise has pointed out, that didn't work.  Focus group No. 1 told us that that didn't work.  So we have sort of ended up with a shorter explicit language on what we mean by "a variety of fish" which you will see in the second rule on the revised advisory.

    In relation to the second recommendation which was to work with other federal and state agencies, you have obviously heard what our approach has been to that.  There has been a very close collaboration between FDA and EPA to develop this singly joint advisory.

    Obviously, that has led to some changes in the advisory which we will go over in more detail tomorrow, but integrating the issue of the message of whether consumers can essentially eat the FDA allowance and then eat the EPA allowance.  There was some concern that we had that that was part of the message that was getting out there so it was critical to try to meld these together.

    In so doing, not only have we just simply merged the language, but we have also gone to pains to try to point out that, under certain circumstances, once you have eaten a sports-caught or recreationally caught ration, that is it for the week.  You are done--which I don't think was coming through before.

    Part of the recommendation was to interact

with states.  We have done that through the stakeholder meetings and through our close involvement with EPA who interact with the states on a very regular basis.

    So the third recommendation was to publish a quantitative exposure assessment used to develop the advisory.  You are going to hear a lot more about that in the next presentation from Clark Carrington.  This was developed in the latter part of 2002, 2003.  It was presented publicly as a poster in March, 2003.

    Following that, and following our initial dialogues with EPA, we all felt that it was critical to get this peer-reviewed which was done through the mechanism that you will hear from Rita Schoeny in the next presentation just exactly how that peer review was conducted and what the results of that were and how we reacted.

    Following the peer review and following some new data that I am going to come to in just a minute, the exposure assessment was further revised so that we have now ended up with another iteration

of the exposure assessment that has really taken into account some new data, which I am going to share with you in a minute, and the comments from the peer review.

    The fourth recommendation was to develop specific recommendations for canned tuna.  As you all know, canned tuna comprises two principle types, albacore or white, and light.  We certainly learned--we knew that canned tuna was one of the most frequently consumed fish in the United States.

    As you will hear from the exposure assessments, we have assessed different scenarios addressing tuna, albacore tuna, specifically, in terms of running scenarios where it is in or it is out and the impact of that on what it does to overall mercury levels.

    To try to get a sense as to what the effect,  through exposure-assessment analysis, of including albacore in your diet, limiting albacore in your diet, taking it out.  Part of this process was to develop new data on levels of mercury in canned tuna.  There were some comments that the

data that we were using was not particularly current.

    As you will see in a subsequent slide, we have addressed that.  As I said, I am going to share those data with you in a minute.

    In the original 2001 advisory, there was no specific mention of tuna.  We took the recommendation that there was a need to do that because it is one of the most frequently consumed fish and have addressed that by including a specific statement regarding canned tuna.  I don't want to go into how we ended up there--you will hear more about that tomorrow--but we ended up with a Q&A approach and we now have a question and an answer regarding canned tuna.

    The fifth point was to address children more comprehensively in the advisory.  As I pointed out when I summarized FDA's advisory and EPA's advisory, there were some differences there in the way we addressed children.  What we did in our workgroup, or our joint workgroup, was we really determined that there was no scientific consensus

to define a specific age or weight of a child in a revised advisory in terms of what constitutes a child by age, what constitutes a child by weight.

    The lack of scientific consensus on the that really limited us in terms of what we felt we should say in an advisory that was going to be a national advisory.  However, what we have done in the revised advisory is put more emphasis on children in this joint advisory than there was in the previous FDA advisory.

    Young children are now mentioned specifically in the title as the target group of the advisory whereas previously they were not.  They are now mentioned more frequently in the text.  Previously, they were really mentioned only in the context of prudence in relation to avoiding certain types of fish, as in the third sub-bullet there.  It is no longer limited to the "Do Not Eat" list which is really where we were in the 2001 advisory and a statement has been added, based on this lack of consensus that I discussed in the first bullet, indicating that children should eat less than the

12 ounces because they are smaller.

    Finally, in terms of increasing monitoring of methylmercury to include levels in fish and the use of human biomarkers--I am not going to go into the human biomarkers issue.  As you all probably know, there was data published recently from the NHANES study going back to the 1999-2000 NHANES study.  I know that was presented extensively at the last Food Advisory Committee meeting.

    But we at FDA have done is we have embarked on two new assignments to measure mercury levels in fish in the U.S.  These were completed in the latter part of 2003.  These two

assignments--the first one was to look at twelve different species of fish with a total of 224 samples.  The second assignment was focused exclusively on canned tuna.  That comprised 170 samples of albacore or white and 119 samples of light tuna.  I am going to expand on this in subsequent slides.

    This is a list of the twelve fish that were part of the general assignment.  These were

chosen because they were fish that we had limited amount of data on, fish that we felt that we just needed more information about.  I am not going to go into detail here, but you can see that there was a wide range of fish that really wasn't focused on the fish that were most frequently consumed.  That was not the purpose of this.  It was to fill in data gaps.  This assignment was initiated in 2002 and completed earlier this year.

    The assignment was set up so that it tested a mix of fresh, refrigerated and frozen fish.  Approximately one-third were to be taken from domestic imports.  These were from the places listed there, Baltimore, Chicago, Florida, Los Angeles, New York, San Francisco, Seattle and the Southwest.  So it was pretty geographically diverse.

    The second part of this assignment which constituted about two-thirds of the samples were from domestic samples.  These were, again, geographically diverse between Atlanta, Florida, Los Angeles, New Orleans, New England, New York and

San Francisco.

    This slide summarizes the data that we obtained from this study.  What I have done here is to, on the same slide, put the old data and the new data.  Let me explain what I am talking about here.

    The new data, first of all, over on the right, is the data that was obtained from the assignment that I have just described to you.  So that is the 224 fish.  The old data are the same fish, the same twelve species.  These data are from our website, which are a composite of older FDA data, some National Marine Fisheries data, and essentially are a composite of data that were used in the early exposure assessment and are the data that are currently on our website which will be subsequently updated with the new data.

    But I wanted to just put these up there for comparison.  I am not going to go through each one line-by-line but I want to point out a couple of things.  Some of these have got asterisks on them which indicate that there are questions about species in these old data that we just don't have

information about.

    I want to illustrate this with tilefish.  The new assignment in tilefish was specifically oriented towards golden tilefish.  In this, we ended up with a mean of 0.208.  Here is the range, 0.05 to 1.1, and a total of twenty samples.  The old tilefish data, which was 60 samples, there is no data on where these fish came from.  There is no data on what species they were.  But, clearly, they were very different.  We had a mean here of 1.45 with a very different range, going much higher.  This is largely why tilefish is in the "Do Not Eat," the "avoid" list.

    What these data tell us is that there is a need for more testing to find out about tilefish.  Is this a change in trend of tilefish or does golden tilefish not represent the tilefish in the United States.  We are currently setting up further assignments to ask questions specifically in relation to more data that is needed to augment these new data over here on the right.

    The same applies, for example, to croaker.

We don't see the big differences in the mean--I beg your pardon; yes, we do--0.28 to 0.054.  Grouper, 0.27 to 0.569.  The n's here are not very large and, really, what this is doing is it is asking more questions.  But we don't know anything about the species of croaker that were looked at in this first set.  We do know these are Atlantic croaker.  Is that important?  Is that an explanation?  Don't know.

    These data are available in your packet so you can study it in more detail if you wish.  But I don't propose to spend any more time going over the numbers here, but simply to point out that some of these differences between the old and the new data may well be related to sampling, geographic questions that we need to address further.

    The second major assignment was the tuna assignment.  The tuna assignment which was initiated in the late summer, mid to late summer this year, looked at approximately 75 percent major brands of tuna, 25 percent store brands, local brands and other brands.  The goal here was that

the sampling would be representative of the volume and type of major and local brands and in packing medium.  So it was water, broth, oil that were available in the areas in which the testing was done.

    This was a market-basket type study so these were just cans off the shelf, so to speak.  But, as you can see, we were shooting for 75 percent major brands, 25 percent of the other type.  Again, this was spread around the country; Los Angeles, San Franciso, Seattle, Chicago, Dallas, New England, New York and Florida.

    This slide shows the data that we obtained from that.  Again, what I have tried to do is to compare old and new data.  Previously, our old data for canned tuna was all lumped together as canned tuna with 248 samples with an overall mean of 0.17.  So this doesn't differentiate between white and light.

    Over here we have the new data, again 170 new white-tuna samples and 119 light-tuna samples.  The old data, so to speak, had a paucity of white

tuna, albacore, of only 17 samples.  Here you can see that we have essentially the mean of 0.29 here, 0.358 there.  Obviously, there is a huge difference in between those two which may to explain this.

    The highest that we found in the albacore was 0.85.  The light tuna--we already had 225 samples of light tuna in the dataset.  We now have 119 more.  Really, you can essentially say that the light tuna has not changed here and the ranges really are not that different either.

    DR. DWYER:  Which is the main kind on the market, David?

    DR. ACHESON:  The main kind that is consumed?  Light.  I don't have the specific split in my head, but I can certainly get that for you for tomorrow if you want to know as to what proportion of the market share goes to albacore and light.  But there is more light consumed than albacore.

    So those are the new data from 2003.  Again, we are not stopping there.  There is a new assignment for 2004 that is about to be initiated

that is going to continue to do this.  We are going to continue to test canned tuna and light tuna along with all the other species plus some.  Actually, the new 2004 assignment for the other species is going to be expanded to fifteen species including a lot more focus on the ones we need more data on, like the tilefish.

    So that concludes, really, the response to the recommendations and the new data.  I just really want to end this presentation by just reiterating our question to the committee which is really broken down into three main bullets.  I will just read these: "Given the enormous interest and expectations from all perspectives in this issue, the one important point that we believe we all agree on is that we move forward and begin our education program."

    Certainly FDA and EPA feel very strongly that we want to embark on getting the word out.  That is what we really want to try to do at this point.  Clearly, as we move forward on this, new science will come along, as has already been

mentioned by Denise, and that will cause us to review and to change and to rethink, but we want to move forward.

    This point is really brought out in the second one, "As we learn more from scientific findings, population demographics, et cetera, and receive results from the education efforts on consumer behavior, we need to refine the approach."  We learned some very important lessons with our focus-group testing.  That was extremely valuable and that really illustrated to me, and this is the first time I have really been involved in focus-group testing from a personal perspective, was, wow; that didn't work.  I understood it but the general public didn't.  So we tried to fix it.

    Then, "We believe that this activity is best conducted concurrently with an outreach and education program, "hat is it in the interest of public health that we get on with this as soon as possible.  Really, what we are looking for is the committee's concurrence and their feedback and comment on what we have done, where we have ended

up, but with the goal of trying to move this forward as quickly as possible, again, with our focus on a scientifically sound, simple and public-health-oriented message.

    With that, my presentation is completed and I think I am open for questions.  Is that right, Dr. Miller?

    DR. MILLER:  That's fine, David.  Thank you.

Questions of Clarification

    DR. MILLER:  Dr. Aller?

    DR. ALLER:  Marion Aller.  On the mercury testing in levels of fish, you mentioned that you were collecting species information on tilefish.  What about the croaker and the grouper, on the new data.  Was that information collected?

    DR. ACHESON:  Yes; it was.  The croaker were Atlantic.  The grouper, there was a wide range of grouper that could have been in there.  I don't have those data with me as to exactly what they were.  But this was very clear to us that we need

species information as we move forward.  Just

simply knowing that they are a croaker group or tilefish isn't enough.

    DR. ALLER:  Thank you.

    DR. DURST:  Durst.  I just have a question on the tables with the analytical data; for example, the next to  the last slide that you have.  None of the tables that I see have any indication of what that mean is.  I assume it is parts per million, or micrograms per gram, or--

    DR. ACHESON:  Oh; I do apologize.  Yes; it is parts per million.

    DR. DURST:  Then I am curious when you have, for example, 225 datapoints, why the range has ND as the low limit on some of these.

    DR. ACHESON:  It is non-detected.

    DR. DURST:  There must be a lower limit of detection?

    DR. ACHESON:  It is essentially the point of detection of the assay.  The other one that says 0.00, I just simply dropped a number off to make it down to just two numbers, two decimal places.

    DR. DURST:  Okay.  Again, speaking as an

analytical chemist, that looks a little curious when you see something like that without a number or some indication of why there isn't a number there.

    DR. ACHESON:  I should point out that much of that data, 225, goes way back in time and were probably a mix of different analytical method some of which were probably less robust, less accurate, whatever--I don't know the specifics--than current methodologies.  It is composite data from a variety of sources over a number of years.

    DR. MILLER:  Dr. Waslien?

    DR. WASLIEN:  When I look at these, I wonder something about frequency of distribution.  Is there one value that is way out there all by itself and, if so, when you do your analysis, you can get batch number and find out if that is coming from one particular area.  That is the chemist or research person in me.

    But I also prefer medians over means.  If you have got one value that is very, very high and you have got all the rest low values, it gets

distorted by a mean.  I just don't know if maybe the means and the medians are the same, so it doesn't matter.  But I do worry that there is one value out there that is your one value and that might be next to a contamination site, whatever.

    DR. ACHESON:  That is a very good point.  One of the reasons that I stuck with means is because all the old data were expressed in means.  For example, with the tilefish, we tried to get back to the original data.  Some of it goes back to the 1970s from the National Marine Fisheries and it wasn't available.

    So you are absolutely right.  This is essentially a summary and the distribution is as important as what the overall mean and the range yes.  Yes.

    DR. MILLER:  Dr. Dwyer.

    DR. DWYER:  Thank you, Dr. Miller.  Thanks for an interesting presentation, David.  A couple of questions.  First of all, will the old values be included and merged with the new and, if so, what standard will you use for throwing values out--for

instance, you said the methods might have changed.

    Secondly, how will the data be transferred and used so that we can get exposure estimates from NHANES and other large-scale surveys?

    DR. ACHESON:  First of all, to address your question of whether the data will be merged.  Certainly, when we don't have specific data that is old, the feeling is to try to replace it with newer information, especially, to reiterate the tilefish issue where we don't know geographic distribution, we don't know species, that might be relevant, clearly, as we develop those data, we want to replace the old data with the new data.

    Where we have more recent, and we haven't defined what we mean by recent, but more recent data that is from FDA labs where we have the specific numbers, then, yes, I think we would attempt to pool it.

    The second part of your question in terms of how will these data be used to NHANES.  I am not quite sure where you were going with that.  Can you clarify what you are--

    DR. DWYER:  Is it going in any database that is accessible to people so that you can do the analysis.

    DR. ACHESON:  Oh; absolutely.

    DR. DWYER:  Or is it hidden someplace in FDA where nobody can find it?

    DR. ACHESON:  The old data is already up on the website.  Our intention is to update the old data, include the new data on the website.  Certainly, we have already passed it out to interested parties who have requested through FOIA, the specific values.  So it is going to be available.

    DR. DWYER:  Just one final question, since the four baddie fish, shark, kingfish and so

forth--are we confident in those values or are they all old values, too?  How old are the new values for those other fish that you didn't analyze, the ones that are supposed to be high in methylmercury?

    DR. ACHESON:  There is more recent data on those that I am aware of.

    DR. DWYER:  What does that mean; 1980,


    DR. ACHESON:  No; 1990s.  I can get you specifics if you need that on what data there is for those other ones.  There is also a lot more of it than the tilefish.  I think that it is a much clearer picture with shark and swordfish and king mackerel than it is with the tilefish.  But you raise a good point, because the data is newer.

    DR. DWYER:  Can I ask one more question, Dr. Miller, and then I will be quiet.

    DR. MILLER:  Of course.

    DR. DWYER:  I think that, for many foods, that sometimes industry produces their own data and, as long as it was done using standard techniques and the sampling frame was appropriate and so forth, these values are accepted--at least, I think, in the USDA, they are accepted if they meet certain criteria.

    So, it is a private-sector effort that has public good coming from it because it all goes into the database.  It sounds like this is all our tax dollars going to find out methylmercury in these

various fish.  Is there any effort on the part of the industry to produce similar data on methylmercury in these various fish?

    DR. ACHESON:  I guess industry would have to speak specifically to that, but I do know that industry have generated data on certain types of fish which they have offered to share with us.  Obviously, if it were to be included, we would have to be sure that the methods matched up and that it was scientifically sound.  Having accepted those criteria, we would certainly be interested in seeing it

    DR. MILLER:  Dr. Archer?

    DR. ARCHER:  Doug Archer.  It may be a minor point, but I noticed that, in some of your slides and in some of your explanatory text, you sort of used mercury as methylmercury.  Is that intended?

    DR. ACHESON:  The advisory is oriented towards methylmercury.  The testing in the fish is actually mercury.  It is total mercury that is tested for in fish.  You are looking like I haven't

answered your question.

    DR. ARCHER:  You have answered my question.  I don't have the background of having been here at the time so I guess my next question would be why.

    DR. ACHESON:  The explanation for that is methylmercury is the form of the mercury that we are concerned about in fish with regard to the impacts on neurological development.  Fish are the primary source of methylmercury.  Virtually all of the mercury in fish is methyl.  The methods for measuring total mercury are simpler than methylmercury.

    So, if you measure total mercury in fish, you have essentially got the methylmercury because it is virtually all methylmercury.  So, arguably, the total mercury is a little bit higher, but very little.

    DR. MILLER:  Dr. Aposhian.

    DR. APOSHIAN:  I think it is appropriate to compliment the FDA and the EPA with coming out with a joint advisory.  The White House Conference

of 1999 was called specifically to try to get three government agencies to agree.  For the FDA and EPA to do this, I think is quite remarkable and I think they should be complimented for it.  I don't compliment people very often.

    I would also like to say I thought the two speakers did a very, very good job.

    DR. ACHESON:  Thank you.

    DR. APOSHIAN:  But I also have a question, namely, will we, at some time, have the opportunity to make suggestions or criticize this advisory or is that the appropriate thing to do now, or what?

    DR. MILLER:  Why don't we wait until we get to the advisory.

    DR. APOSHIAN:  Okay.

    DR. MILLER:  It is kind of hard to criticize something that we haven't theoretically dealt with yet.

    DR. APOSHIAN:  Okay.

    DR. MILLER:  Although we could do it.

    DR. APOSHIAN:  We can wait.

    DR. MILLER:  Dr. Russell

    DR. RUSSELL:  There is a fairly wide range when you look at the tuna with regard to the levels.  Is there any geographical explanation for that?  That goes for both the white and the light tuna.

    DR. ACHESON:  We have not specifically looked at that.  There may be data out there that gets at that but, obviously, when you pull a can of tuna off a supermarket in Washington versus one in Seattle, there is no way of telling from that can where that fish came from.  So I don't have the answer to your question.

    It may be that there are other data sources that could get at the answer to that in terms of whether there are geographic differences.  I would suspect that they are not huge because, essentially, tuna are from the ocean.  The levels in the ocean are fairly even.

    DR. MILLER:  Rob, does that answer your question?

    DR. RUSSELL:  Yes.  I guess I wondered, do you have levels for fresh tuna as well from various


    DR. ACHESON:  No; not in relation to recent testing.  Fresh tuna is going to be part of our 2004 assignment.

    DR. ACHOLONU:  Could you slow that slide again on mercury levels in various fish?  I would like to see it again, please.

    DR. ACHESON:  Okay.

    DR. ACHOLONU:  There you have trout, fresh-water.  And you have the mean as 0.42.  Then you have the range, 1.22 maximum.  Yet, under the next column, you have NA.  How were you able to get the mean, the trout, fresh-water?

    DR. ACHESON:  The trout, fresh-water, is the old data that is composite from a variety of sources.  The next row, which is farm-raised trout, are the new data.  So we did not have any old data on farm-raised trout specifically.

    DR. ACHOLONU:  You got that from somewhere.  What does N signify?

    DR. ACHESON:  Not available.

    DR. ACHOLONU:  So you don't have any

number available but you have the mean for trout in fresh water.  How did you get the mean?  That is what I am trying to--

    DR. ACHESON:  I am a little--

    DR. HEIMBURGER:  It must have been reported by someone else.

    DR. ACHOLONU:  So you got it from somewhere else?

    DR. ACHESON:  Yes.

    DR. ACHOLONU:  That is what I am trying to find out.

    DR. ACHESON:  Okay.  Like I tried to point out, the old data are from a variety of sources.  Some of it, we had the specific numbers on and some of we didn't, which is one of the problems with the tilefish is we did not have the specific numbers, which is why we couldn't calculate medians and a variety of other dilemmas.

    DR. ACHOLONU:  Thank you.

    DR. MILLER:  Dr. Downer.

    DR. DOWNER:  Goulda Downer.  I, too, want to commend the two groups in working together and I

am really very interested to get a little bit more information on the discussions you had with the focus groups.  In preparation for this meeting, I asked just about every one of my patients last month, who would usually have tuna and soup for lunch, if they were still having it.

    And everybody, 100 percent, said they no longer ate tuna.  I don't know if you are working with the media to make sure the message that comes out to the public is not one of, "Don't eat the tuna," but, "You can go ahead and have some."  People were just so scared.  They are saying to me, "Dr. Downer, I don't know what to eat and what not to eat, so I am just not going to bother."  These are not women of childbearing age, most of them, I might add.  They are women past childbearing age but they are thinking, "There may be some additional implications for me."

    DR. ACHESON:  Right.

    DR. DOWNER:  I am just curious to find out from you what kind of educational materials you have started to develop and how is that going?

    DR. MILLER:  Let me just respond.  As far as I can see, we haven't gotten to the advisory yet.  Also, am I incorrect in saying that the distribution of the advisory hasn't taken place?

    DR. ACHESON:  That's correct.  The advisory is a draft.  It is available for people to look at as a draft, but it is not--subject to, obviously this meeting and other comments, it could change.

    One of the presentations tomorrow, the first one by Marjorie Davidson, is going to get much more into the focus groups.  If it is okay to defer questions related to that until tomorrow, I think that might be--

    DR. MILLER:  Why don't we wait until the presentation tomorrow.

    Dr. Dwyer?

    DR. DWYER:  Trouble again.  Back to Dr. Russell's question.  I just wanted to get this into the record and get somebody to answer it, and maybe it is you, Dr. Acheson.  I think two years ago we heard from the halibut people about--I think it was

halibut--huge differences between the Alaskan and New Zealand, I think it was, and other parts of the ocean.  It was basically that there were certain parts of the ocean where the fish had almost no methylmercury in them even though they were the same species as fish in other parts of the ocean.

    Is it because the tuna are--why wouldn't that also be true of tuna?  I was just following up on Dr. Russell's--is it because the halibut live in different places or what?  I just don't know very much about fish.

    DR. ACHESON:  I am not qualified to answer that question in relation to what the true data is on tuna.  I speculated that tuna may not be that different, coming out of the same ocean, but I certainly could be wrong and subject to correction.  I will try to find out the answer to that question for you in terms of what we know of the distributions of the levels in tuna as it relates to different parts of the--geographic variation.  I don't know.  It was a speculative answer and I qualified as such.

    DR. DWYER:  Am I incorrect?  Didn't we hear a whole bunch of impassioned people from New Zealand and Alaska saying that they had been maligned because their fish didn't have any in it?

    DR. WASLIEN:  Wasn't it salmon, which spends a lot of its life out of the ocean.

    DR. MILLER:  There are a lot of fish that are called by the same name that are different species, actually.  That is part of the problem.  So some fish that are called halibut, two fish, both called halibut, could be two quite different species.  There is a commercial advantage to calling the fish by these names.

    DR. ARCHER:  If I may add just one other point in relation to that, the one thing that we do know is that the size of the fish has an impact on the amount of mercury in it.  Certainly, that can affect, depending on how big the fish is that goes in the 6-ounce can, would have an impact, outside of geography.

    DR. MILLER:  Dr. Callery.

    DR. CALLERY:  This is a follow up

analytical question about the tuna.  The light tuna is substantially lower than the albacore but I believe the albacore is often, or usually, packed in water and the light in packed in oil?

    DR. DOWNER:  Goulda Downer.  You can get them in both, water or oil.

    DR. CALLERY:  You can get them in both but I am wondering why, if there is a known reason why, the light tuna would be half the amount as the albacore and just wondering out loud if, in fact, the mercury is in the oil.

    DR. ACHESON:  No.  I think the differences are related to the size of the fish.  Bigger fish generally have more methylmercury than smaller fish, and the smaller fish are typically the ones that wind up being as light tuna and the larger fish as white or albacore.  It is not as straightforward as that but, in essence, that is the explanation.

    DR. MILLER:  Dr. Lund?

    DR. LUND:  Daryl Lund.  A couple of questions.  One is the tuna were all canned; is

that correct?

    DR. ACHESON:  The tuna in the new data were all canned; correct.

    DR. LUND:  So you could get from the can code--if you recorded those, you could get at least some location information about where those tuna might have come from, if you recorded the can code on the lid.

    DR. ACHESON:  Theoretically, yes; if you recorded the can code and then went back to the manufacturer, I am sure you gather more information about that.  We have not done that.

    DR. LUND:  Then, secondly, on the other fish, under the new data, those were all processed in some way?  Are they frozen or are they fresh out of the ocean or out of the lake or out of wherever you took them?

    DR. ACHESON:  They are a mixture.  They are a mixture of fresh, refrigerated and frozen.

    DR. LUND:  Okay.  I understand.

    DR. ACHESON:  I don't have a breakdown as to how much fresh, but they were a mixture.

    DR. LUND:  Thank you.

    DR. MILLER:  Dr. Krinsky?

    DR. KRINSKY:  Norman Krinsky.  May I just ask why salmon is not included on your list?

    DR. ACHESON:  On the list of the fish that were tested?

    DR. KRINSKY:  Yes.

    DR. ACHESON:  That list of twelve was developed based on the fish that we had a paucity of data on or there was some specific need to get more information about, and it was essentially a resource-limitation issue and salmon was not a fish that we needed more data on last year.  So that list is not meant to include all the fish that we are interested in.  It was a subset that we tested with new data in 2002-2003.

    DR. MILLER:  Dr. Heimburger?

    DR. HEIMBURGER:  Heimburger.  A follow up to that.  So the eventual posting of data will be a comprehensive list including salmon and other fish that were not tested in this time period.

    DR. ACHESON:  Absolutely.  The old dataset

that is up there is a piece that I took out of what is already posted which includes salmon and a whole range of fish.  I just, for purposes of comparison, and not to sort of have a slide that was uncomprehensible, stuck to the comparison with the new data.  That is the reason.  Please don't get the impression that if they are not on that list, we don't care about them.  We do.

    DR. MILLER:  Any other questions?  Some of these issues can be re-raised again at the end of the session.  Right now, it is supposed to be time for a break.  Why don't we just take minutes.


    DR. MILLER:  The next series of presentations are on Exposure Assessment and Peer Review.  Rita Shoeny from EPA.

Exposure Assessment and Peer Review

Peer Review

    DR. SCHOENY:  Thank you.  We are going to spend the rest of the afternoon, I believe, talking about the exposure assessment.  I am going to begin the discussion with some brief remarks about the

peer review that was conducted by EPA and FDA on this exposure assessment.

    The main reason why the exposure assessment was done was in response to the recommendations that were made by the Food Advisory Committee in 2002.  That was recommendations on our fish advice, which Dr. Acheson has already addressed.

    There were at least two of your recommendations that were specifically addressed in the exposure assessment.  The first one, of course, was to do an exposure assessment.  But, as was also pointed out, the exposure assessment deals with some scenarios for fish consumption limiting either the amount of fish or the types of fish consumed, and that is where we specifically talk about canned tuna and some other fish varieties as well.

    We had available to us at the beginning of our discussions about the advisory the poster presentation and an earlier paper that were put together by Drs. Bolger and Carrington.  We began using that exposure assessment as a tool to help

focus our discussions on what could reasonable go into an advisory.

    We also discussed some of our conclusions with the stakeholder meeting.  At the stakeholder meetings, we presented some of our early conclusions about the exposure assessment.  At the time that we were addressing the stakeholders, we felt, as a group of scientists, that the model that is described in the exposure assessment closely predicted the NHANES data that you have referred to already.  In particular, we were referring to the publication by some of the NHANES investigators that remarked that about 8 percent of women of childbearing age in the United States are at a blood level for mercury either at the reference dose that has been published by EPA or in excess of that reference dose.

    We felt that the fish scenario model, the fish exposure assessment that had been done by FDA, closely approximated the results that were presented in the NHANES paper.  We thus felt that this was a useful tool in establishing the

scientific background for specific advice.  We felt that it was a way to inform our risk-management decisions and we also told our stakeholders at that time that we were submitting the exposure assessment to peer review.

    What was reviewed was a poster presentation that was publicly given at the Society of Toxicology Meeting last year.  We provided, as well, an earlier publication by Drs. Carrington and Bolger which described the development of the model that was used and modified somewhat in the poster presentation.

    The poster abstract has been published.  It has been published in The Toxicologist.  We also have the citation here for the earlier paper.

    The review was done using an existing EPA peer-review contract.  It is what we call a letter review or a pass-around review.  This is a situation where we send out the documents to be reviewed to the reviewers.  They submit their comments back in writing.  There is a written report.  This is as opposed to a public meeting

such as the Food Advisory Committee or a Science Advisory Board meetings in EPA.

    EPA and FDA scientists describe the expertise that we felt was required to provide an appropriate review of this assessment.  The contractor selected three reviewers.  EPA, as the holder of the contract, approved the entire list of reviewers as having the requisite credentials.  Then the contractor provided the material to the reviewers, collected the written comments from the reviewers and then, ultimately, compiled the peer-review report.

    EPA and FDA wrote the charge to the reviewers jointly and I will go over this briefly.  As with any sort of review that we do of a piece of work that is going to be available to the public, we wanted to know if the document was logically written, if the arguments were clear and concise and presented in an understandable manner.

    We certainly wanted to know of all the appropriate literature had been cited and if there were other publications that we should be

considering.  As this exposure assessment is dependent upon a model, we wanted to ensure that the model was clearly described and, as there were modifications made between the time of the publication of the model and the poster and subsequently, what we were requesting is comment on the modifications that were made.  Were they supportable by existing data?

    Here are some of the modifications that were made; expansion of fish categories, fitted distributions, addition of an amount of mercury to the predicted blood levels to account for exposure that was not fish related--for example, from amalgams, medications and so forth.

    A keystone of the inputs are the data from the continuing study of food intake by individuals.  The exposure assessment used the data from 1989 through 1991.  There are more recent data--that is, from 1994 through 1996--or I should say the FDA chose to use the three-day survey data that were somewhat older.  We asked for comments on that use of data and also comments on adjustments made to

compensate for likely underreporting in this short time-frame sample.

    We also requested comment on the range of ages that were used for women of childbearing age and for children.

    We asked of the fish scenarios were logically and clearly described.  We had two questions, actually, asking whether or not some cutoff points that were set categorizing the fish as either low, medium or high mercury for purposes of the fish scenarios, whether this were appropriate characterizations and cutoffs.

    What we received back in response was a 37-page report that we felt was responsive to the charge questions and also addressed a number of additional issues.  The contractor compiled the report.  There was no summary--that is to say, they were not looking, the contractor, themselves, for consensus across the reviewers.  In our response to the reviewers, however, we do some summarization of review comments, particularly as they addressed the charge questions.  I will go over those briefly.

    We were gratified to note that they thought the documents were well written and clear.  There were some suggestions of additional literature that were made.  The model was well described in the published paper.  The reviewers felt that some of the modifications and adjustments were made and others they critiqued; for example, the adjustment of the three-day data for long-term exposure.  Dr. Carrington will talk a bit more about some of these adjustments.

    Two of our reviewers felt that the use of the older three-day data were more appropriate than the use of the two-day survey data.  All the reviewers commented that we were likely to be underestimating the number of fish consumers using the CSFII data.

    The definition of women of childbearing age was considered to be a policy choice by all of our reviewers.  That having been said, though, they at least felt that these were reasonable age ranges to use for the scenarios and exposure assessment.

    All of the reviewers suggested

improvements in our description of the fish scenarios.  Many reviewers suggested additional scenarios to be run.  Some of these have been run.  All of the reviewers thought that cutoff and categorization of the fish species as high, medium or low mercury was appropriate.

    EPA and FDA provided a written response to the reviewers, and I understand you have that in your package.  The review report, itself, as well as our responses, are available on the EPA and FDA websites.  As you look at our response to reviewers, you will see that this describes revisions to the assessments and some of those will be subject of Dr. Carrington's remarks.

    We also feel it necessary to comment on areas where we disagreed with the reviewers to lay out our position on that disagreement.  You will find that in the response to the reviewers as well.  These were very voluminously and comprehensively written reviews for which we were very grateful.  However, some of the comments that were made we felt were outside the scope of the assessment that

we were doing now.

    There were a lot of comments made by our reviewers in terms of what additional items one should address for fish advice of various sorts.  Should we be talking about other persistent bioaccumulative toxicants in this fish advice.  You will see that we addressed some of those.  And we also said, yes, some of these comments that you have made we think are fertile areas for future work.

    There were a lot of recommendations made in terms of, well, if you spent the next nine months, you could also model thus and such.  You could make these enhancements and so forth.  Given the fact that we think it is really useful to put out some advice in a relatively short time frame, we have not ignored those comments but we consider those as being some things to approach for the future.

    The exposure assessment has been modified.  Drs. Bolger and Carrington have been refining the exposure assessments since they first gave their

poster presentation.  They have been using newer data on tissue concentrates of mercury in various fish species, as we mentioned earlier this afternoon, and they have also made some changes.  Again, Dr. Carrington will explain.  These include more categories of fish, some correction for food preparation, changing some of our parameters to reflect the NHANES data more closely, variations in consumer fish choice.  I think Dr. Carrington is going to go over this.

    Some of the scenarios were changed to reflect both the limit on the amount of fish consumed and the type of fish consumed and combinations of both.  Our reviewers commented to us that there are many areas in the country or many fish consumers where one species or a limited number of species are being consumed.  That is something we should talk about.

    Some of these consumption scenarios addressed.  And also the body-weight scaling was changed.

    At this point, I will stop and ask if

there are any questions for clarification on the peer review.

    DR. MILLER:  We can come back to this later on.  I guess next is Clark Carrington.  Why don't we move on to the Exposure Assessment.

Exposure Assessment

    DR. CARRINGTON:  I guess first I will warn you that there is a lot of stuff that I am not going to explain.  I am going to start out talking about my model and I am mainly going to talk about the changes that were made since the last publication.  If you are not familiar with the model, there is a lot that is going to go unexplained.

    I do have another set of slides that has some technical issues that I could go through, but I will only go through those if somebody asks me to.

    After going through the model and the changes I made, I am going to briefly compare the output of the model to the NHANES survey as a validation exercise.  Finally, I am going to talk

about some simulations for some putative advisories that can be used to evaluate what the expected impact of the advisories is, assuming that they are followed as specified.

    I have broken down the changes that were made to the model.  These are all changes that were made in basically the last year.  I have actually be working on--I have had some form of this model since 1994 when I actually published an early version of this model.  So I have been working on it for a long time.  I have been working on it more lately, but, still, this has been an iterative development process and these are the most recent changes.

    First of all, the number of categories actually stands at 41.  I think Rita said was expanded from 24 to 28, which was actually--that was the expansion going from the paper to the poster.  But, since the poster, I have actually expanded it to 41.

    A lot of that expansion occurred as a result of getting new data.  But I also expanded it

in order to try and make it match up with the NHANES categories because I think, in the future, I am going to try and use more of the NHANES data.  So I am trying to get the concentration data to match up with the NHANES consumption categories as best I can.

    Finally, the biggest change in the model, at least in terms of what the average predicted level of exposure is, came about from introducing a correction factor for weight loss during cooking.  This, on the average, is about a 20 percent factor so it resulted in a 20 percent increase in predicted blood levels.

    Then I also made some changes on the part of the model which predicts consumption frequency for each individual in the survey.  I am still using the same equation that I used before which is an exponential function which adjusts the frequency observed in the CSFII survey to generate the predicted long-term frequency.

    However, I have changed the parameters of the equation so now it is consistent with the

consumption frequency observed in NHANES.  Then I also increased the range of my estimated number of total seafood consumers to make it consistent with the number of consumers reported in the NHANES seafood survey.

    Finally, I made some changes in the way the model selects species for each individual

for--I have changed the way the model picks the species that each individual consumes over the course of a year.  In the previous version of the model, I treated this value as a population variable so that somewhere between 20 and 80 percent of the time, I would use the CSFII survey to select the fish and then the rest would come from a market-share distribution.

    But, regardless of where it was between 20 and 80 percent, every individual in the survey had the same number.  I have changed that so that is now an individual variable so that each person in the survey has a different number.  Some people are eating the same fish all the time and other ones aren't and I am using a different range which is

based on the NHANES survey.

    To sort of make a very long story short, this is what the output of the model currently looks like.  It is still a pretty good fit but, instead of slightly underpredicting blood levels, now I am slightly overpredicting blood levels.  It is off, depending on what percentile you look at, by about 15 to 30 percent.  It is closer at the high end than it is at the lower percentiles.

    Now, I will go through the advisory scenarios.  I have gotten them broken into three scenarios.  The first category is a series of scenarios where we are just limiting the amount of seafood consumed.  I have got three different scenarios which correspond to limits of 18, 12 and 6 ounces.

    Then I have got a series where we are just limiting which species are consumed without limiting the amount.  That revolves around dividing the fish into different groups.  On the next slide, I will show you what those groups are.  But, the way the simulation runs is that whenever a consumer

encounters a fish that is prohibited, that fish is replaced with another--I guess that item of seafood is replaced with an item from a market-share distribution from the low-mercury group.  So it replaced with a low-mercury fish or shellfish.

    Here are the groups.  There is a high, medium and low group.  The high group basically corresponds to what our current prohibited list is.  Tilefish isn't on the list because I wasn't sure where to put it because we have conflicting data.

    The middle group is basically defined--the low group is largely defined by light tuna; in other words, in order to keep the market share at a reasonable level, we had to keep light tuna in

the--or we felt like we had to keep light tuna in the low-mercury group.  So anything that is significantly higher than light tuna and is not in the prohibited list is in the middle group.

    Where the levels end up with those criteria is anything between about 0.15 and 0.6 ppm is in the middle group.  If it is below that, it is in the low group.  If it is above that, it is in

the high group.

    Finally, we ran some scenarios with combinations of limits on the amount of seafood consumed and limits on the species consumed.  The first one we ran, which basically corresponds to our existing advice, which I am going to refer to as the 12-ounce No High scenario, eliminate the high group and then has a total of 12 ounces from the remaining two groups without any restriction whether that 12 ounces comes from the low or the middle group.

    The next scenario which we are calling the 12-ounce Variety scenario has a 6-ounce limit from the medium group with a 12-ounce total, so up to 6 ounces can come from the middle group and the rest of the 12 ounces has to come from the low group.

    Then the next two scenarios are similar to the 12-ounce Variety scenario except that is an actual reduction in the total amount that can be consumed which corresponds to the amount that is consumed from either albacore in the first case or the middle group in the second case.

    I will just give you an example of how it works for the medium group.  In the 12-6 Medium scenario, if someone eats 6 ounces of medium, then they are done.  So they can't eat any lower.  I guess another example is you could eat 3 ounces of medium and 6 ounces of low.  So, as far as the total is concerned, the medium counts double.

    Then the last scenario is the 12-ounce Low Only scenario where a person can eat 12 ounces and it all has to come from the low group.

    Here are the results of the first series, and I will only briefly--there are a lot of numbers on there and so the most important ones to show or look at--the predicted levels that are above the reference dose, which is given down here, are all going to be in the upper--the 90th percentile and above.  On the other hand, if you have significant reductions in seafood consumption, it is going to show up in the average, which I think is the best number to look at.

    So if you are reducing the average, you are reducing seafood consumption which is not

necessarily something you want to do.  On the other hand, if you are reducing the mercury levels in the 99th percentile, then you are accomplishing your goal.  This last number is the predicted percentage of the population above the RFD.  You will note that a restriction of 18 ounces per week has relatively little effect.  A restriction of 12 ounces has a slight effect, but it is still a relatively small reduction.

Finally, you will see a pretty substantial reduction with 6 ounces per week.

    This is the series with species limits.  The baseline is the same as before.  It is also the same as we have plotted in that graph I showed.  I guess the main conclusion from this is taking just the high group out has next-to-no effect whereas restricting it to the Low Only group has a substantial effect.

    Finally, I have got this list.  I think I have got this graph, but the main thing to point out from this table is these three groups in the middle where you are adjusting the amount coming

from the medium group are not substantially different.  There is a reduction seen in the 12-ounce No High scenario but there is not much of a further reduction in the 12-ounce Variety, Albacore or Medium scenarios but there is a much greater reduction with the Low Only scenario.

    I guess the other thing I would like to show before I move the graphs, which you can't tell from the graphs, is that the numbers in parentheses are confidence intervals and the confidence intervals go up dramatically as you move out to the tail of the distribution.  So the uncertainty estimates at the 99.9th percentile are pretty wide.  That is one of the reasons I am not going out any farther than that.

    Here is a graph which is plotting the median values from the table I just showed you.  This one on the far right is the baseline.  Then there is another group of three here which are all the 12-ounce various combinations of the scenarios that have 12-ounce limits and eliminate the high group.  They are all fairly similar.  You will

notice, they do eliminate most of the area beyond the reference dose but they don't completely eliminate it whereas the 12-ounce Low scenario does, in fact, come pretty close to the goal of getting everybody below the reference dose.

    Just as a summary, numerous revisions have been made to the model since the last time I published a description of it.  Instead of slightly underpredicting NHANES, the model now slightly overpredicts NHANES.  Finally, the percentage of individuals about the reference dose can be reduced by either limiting seafood consumption or by limiting the species consumed, and the degree to which the exposure is reduced is entirely dependent on how aggressive the limits are.

    And I will stop there and take questions.

    DR. MILLER:  Thank you, Dr. Carrington.

Questions of Clarification

    DR. MILLER:  Comments?

    DR. APOSHIAN:  I think we have to more or less view this with a few words of--or a few fears.  The risk assessment and model-making, I think most

people would agree involves a total population.  It does not involve a population that is hypersensitive, in this case to methylmercury, a population that cannot get rid of methylmercury from its cells.

    A perfect example in medicine of this situation with another metal is Wilson's disease or hepatolenticular degeneration where humans die of this because they don't have the mechanism to transport copper outside their cells.  The only way of keeping these people alive, and we can do this now, is by giving them a chelating agent.

    I am passing around to you something.  I would like you to look at the figure of the table on the top right-hand side.  This deals with autism.  I am not getting involved in what is causing autism.  That is not what I want to speak about.  What I want to show you here is that when the first baby hair of autistic children was analyzed for mercury and for methylmercury, particularly, and compared to the hair of normal children, it was very surprising to find the hair

of normal children was much higher as far as mercury content than the hair of autistic children.

    The mercury in the hair is a reflection of the mercury in the blood.  The reflection of the mercury in the blood is whether the mercury gets into the blood.  On top of this, there is a paper coming out very soon, if it is not out already, in which, when they gave a chelating agent, gave a substance that would increase the excretion of mercury, get mercury out of the cells--this was done by very reputable people--when they gave a chelating agent, DMSA, to the autistic children and the normal children, it was found that the autistic children excreted six times more mercury than the normal children.

    Again, I am not saying that autism is caused by mercury, whether this is a result of the autism or the cause of autism is the beside the point.  What I want to point out to you is that here is the first group of people that we know, now, who are unable to excrete methylmercury.  Therefore, the amount of methylmercury in their

tissues is higher and, therefore, is toxic.

    Models don't take this into account.  Risk assessment does not take this into account.  The only thing I want to do is caution this committee to consider that not everyone is a normal human being.  Not everyone can get rid of the mercury.  For those people who cannot get rid of the mercury, and this is just the tip of ice berg--for those people who cannot get rid of mercury, the amount of fish, the amount of methylmercury in the fish that they are going to eat, is going to be very, very crucial.

    I hope you just take this--what I handed out was the datasheet and the reference to the publication.  The title page is attached to it for anyone who wants to look at it.  I think that this is really the first example now of probably a deficit in the transport mechanism, the efflux mechanism, for getting methylmercury out of the cell.

    Thank you.

    DR. MILLER:  Are you arguing that the

levels that should be set for consumption of mercury be based upon these hypersensitive people?

    DR. APOSHIAN:  Are you asking me?

    DR. MILLER:  Yes.

    DR. APOSHIAN:  What I am saying is that we cannot ignore a subset of the population that cannot excrete methylmercury and that most models have to take into account only the majority of the population.

    DR. MILLER:  That's right.  So the question is, if we try to take into account that segment of the population that may be hypersensitive, for one reason or another, how do we do that?

    DR. APOSHIAN:  You have got to issue a word of caution.  You have got to issue a word of caution that--it is sort of like--to me, the methylmercury story is very similar to the alcohol story.  Almost every bottle of wine you buy now will say that excessive consumption of this material may cause a defect or a problem in the child that is going to be born, or have

reproductive effects.

    No one has yet found out, although there is an Indian population that is very susceptible to alcohol, but no one has tried to just point out that certain people in the population are more susceptible to alcohol, pregnant women are more susceptible to alcohol than others.

    I am just saying that we have to take this into account in our warning system.

    DR. DURST:  Durst.  With respect to that, I guess my concern is that, for example, in this autistic case, how do they know they are in that risk group?  Giving a warning, in this particular case, anyway, would not do any good except to frighten a lot of people from eating fish.

    DR. APOSHIAN:  No one says that you shouldn't eat fish.  We are not saying that.  Just like no one says you shouldn't drink wine.  It is like the State of California does have a cautionary statement on many of the products that are sold over the counter.  They don't try to identify the groups but they just point out that there is a


    I know that the fish manufacturers would, obviously, never agree to putting on a can of tuna fish that one should eat this very, very carefully.  I just wanted to make you aware of some of the latest studies being done which really have amazed many of us.  They are very good studies.

    I am sorry I didn't answer your question very well, Sandy, because I don't know the answer to your question about how we take care of everyone.

    DR. MILLER:  I think whatever the outcome is, I think that the process by which these advisories are made can always be modified as data becomes available.  In other words, whatever the presentation, whatever approach is presented by the agency tomorrow can be modified.  But I think that the problem that FDA faces and, in fact, all regulatory agencies face, is how to deal with the bulk of the population versus those parts of the population that may be ultrasensitive and do it in such a way that the positive aspects of the food

involved are not lost.

    DR. APOSHIAN:  With all of your experience with the FDA on this matter, how would you suggest that this--

    DR. MILLER:  I don't know.

    DR. APOSHIAN:  Okay.

    DR. MILLER:  I don't know.  Jean has got an answer.

    MS. HALLORAN:  I have a suggestion.  This is reminding me of how certain products have--there is a sweetener that is a problem for people with a certain metabolic deficiency that has letters that stand for it.  PKU; there we go.  If there is a special group, it could be labeled in a way similar to people with that problem.

    DR. MILLER:  Yes.  But I wouldn't want to see a general recommendation to the population be ignored and base it on the population of ultrasensitive people.

    MS. HALLORAN:  No; my only point would you would could special labeling for that purpose and also having the general advice.

    DR. MILLER:  I agree.  That is an approach.  Dr. Russell?

    DR. RUSSELL:  This is actually not on that particular topic, but in the slides, it says exposure assessment has been revised and expanded.  Your last point was body-weight scaling changed.  Can you explain that, how that was done?

    DR. CARRINGTON:  It actually didn't change since the paper.  That was a change I made right before the paper, as a matter of fact.  I did it primarily to get the model to fit children better.  Anyway, the change was I used body weight to the power of one-third which actually fit not only the adult data better but it also makes the model fit children better.  So it is not straight body weight as a function of dose, it is body weight to the power of a third.

    DR. RUSSELL:  Okay.  Thanks for that clarification.  The other thing was, in the table you have with the high, middle and low mercury concentration groups of fish, I noticed that there are several fish--for example, like fresh tuna,

that we were just told a little while ago that that actually was going to be done in 2005 or something like that but that you don't actually have those figures.

    So I was wondering how you placed those fish in that table.

    DR. CARRINGTON:  We do have fresh tuna data.  In fact, we have lots of it.  That is one of the--

    DR. RUSSELL:  Then I guess we were told that you didn't.

    DR. CARRINGTON:  That was a list of the stuff we had done last year.  Actually, we did other stuff--like, actually, I have gotten FDA data sort of in three parts.  For a long time, we have had shark, swordfish and canned tuna.  Then, actually, one of the differences between the paper and the poster is I got data for maybe 20 species that we had accumulated in the last four years, I believe.  In that data was fresh tuna, so we do have fresh tuna data.  it is just that it is not brand-new data.

    DR. ACHESON:  Dr. Miller, may I just clarify that point?

    DR. MILLER:  Sure.

    DR. ACHESON:  Just to make sure that the committee understands that the data I showed was the new data.  It was the new 2003 data.  We didn't test fresh tuna as part of that assignment.  Therefore, it wasn't included in my data table.  I don't want to imply that if it wasn't on that table that I showed, we don't have data on it.

    DR. CARRINGTON:  Actually, we, in fact, have something like 80 percent of the market covered at this point.  There is still about--I would say the biggest categories we don't have data for--the biggest single category we don't have data for is crabs.  That is something we are still using the National Marine Fisheries Service data.  I guess that would be number one on my list of what we don't have data on.

    DR. MILLER:  Dr. Dwyer?

    DR. DWYER:  Thank you for a very interesting presentation.  I had two questions,

three, actually.  The first was was the model you presented today published in the peer-reviewed literature or is it the old model and are you planning on publishing the new model in the

peer-reviewed literature?

    DR. CARRINGTON:  The answer is--well, the answer to the first question, it is a modified version of the published model so, as it now stands, it has not been published although I think we are planning on doing so; yes.

    DR. DWYER:  You are.

    DR. CARRINGTON:  Yes.  We will; yes.

    DR. DWYER:  Thank you.  The other question was, I guess, to our colleagues at FDA.  It is my impression, and I  may have the name wrong, that the Institute of Medicine of the National Academy of Sciences has just published, I think, several volumes on the whole issue of vaccines and neurodevelopmental and a whole bunch of other hazards.  I wondered if it would be possible to get those volumes and bring them in tomorrow.  I assume that this issue was also reviewed in those volumes

and I would like to know what the Academy says about this issue as well, the issue of reduced levels of mercury in autistic children.

    DR. CARRINGTON:  Is that a question?

    DR. MILLER:  Let's get an answer to that.  Can we get--

    DR. CARRINGTON:  The vaccine issue is inorganic mercury, isn't it?  I guess it is ethyl mercury.  That's right.  Actually, it not that different.  Its pharmacokinetics are different but it is--

    DR. MILLER:  David, do you want to respond to that?

    DR. ACHESON:  That is really the Chairman's prerogative.  Certainly, our goal was to keep this restricted to methylmercury in fish.

    DR. DWYER:  Fair enough.

    DR. ACHESON:  I am sure we could find the data that you want but it is taking us beyond methylmercury in fish.

    DR. MILLER:  But the issue concerning what happens when you have a specially selected

population is--I think that there is a level of data that you need to do this and, if you are going to do something like you did with PKU and asparatame, you need to be able to clearly define what your endpoint is so you can identify which of the children are going to respond inappropriately.

    For the kids with PKU, they are identified at birth.  It is very straightforward which kids are phenylketonuric and which are not.  Autism is a much more complicated issue, I suspect.

    DR. HEIMBURGER:  Heimburger.  I have a question.  Just to follow up exactly on that, unless I am misunderstanding the abstract in this paper that you presented relating to autism, or autistic children, unless I misunderstand autism, it is a diagnosis that is made well after birth.  There is no way to identify, in utero, or reduce prenatal exposures to anything in a child who is destined to be diagnosed as autistic later on.

    Just to clarify, I would think any impact that it might have on any advisory would have to be--there would be no way that I could see to

reduce prenatal exposure to mercury.

    DR. APOSHIAN:  No.  Please understand.  I was not trying to say in any way that autism is caused by mercury.  That is an entirely different question and it is one that is unsolved.  But the point I am bringing up is that, once these children are born, they are going to have diets of fish.  If they can't get rid of the methylmercury, they are just going to be made sicker.  That's all.

    The point I want to make also is we know that about autistic children and studies are just beginning now about other populations.

    DR. MILLER:  Perhaps it would be fair to say that the committee might want to consider a statement such as the agencies ought to be advised to continue reviewing the literature as it changes and make modifications in the advisory where appropriate and to take into account new data.  I think that is the best that you can say for that.

    Dr. Durst?

    DR. DURST:  Durst.  Dr. Carrington, I am curious about your comment that the body mass is to

the third power in your model.  I am not familiar with the model or how many variables you have in it and so on, but I am curious, does  that number, the third power, come from biological or physical relevant matter or is this a number you put in to tweak it to fit the NHANES data.

    DR. CARRINGTON:  I actually got it

from--the data I used to build the model on is a U.K. study which is Sherlock, et al., 1984 in Human Toxicology.  It was 20 subjects with controlled fish diets.  In that paper, they actually--first of all, if you look at the data--I have a slide of this.  Can we pull it up?  Okay.  Somewhere, I have another slide.  I guess the bottom line is that if you plot the blood diet ratio as a function of body weight, there is a correlation there but it is not directly proportional to body weight.  They suggested that the best fit is one-third power which, I believe is--they didn't how they got that.  They just suggested that that was a good number.

    I actually recently went back and analyzed it myself and I got as my optimum power of 0.43; in

other words, a body weight of the power of 0.45 provides the optimum--you can account for the most variance possible out of the total variance as a function of body weight with a body weight to the power of 0.43.

    DR. MILLER:  There is a relationship called metabolic mass which is approximately to the one-third.  I don't remember what the exact number is, but that is a very old concept.  It goes back to the days of people who were dealing with mathematical biology and energy metabolism.

    DR. CARRINGTON:  Also, a lot of it has to do with how the compound is partitioned between fat and blood.

    DR. MILLER:  Metabolic mass is supposed to take that all into account.  The only animal that doesn't fit on the curve is the rabbit and it does if you take its ears off.

    Dr. Lund?

    DR. LUND:  Daryl Lund.  This question about identifying unique subgroups within the population and how you take those into account,

obviously, we are going to gain a great deal more information as individuals have more information on their genetics, on their genes generated.  So I assume--I don't think we need to tell FDA that, in fact, as information becomes available on subsets of the population, that you will take this into account in advising the population as to its diet, or as to limits on its diet.

    Clearly, with the identification of PKU, and once you know, or even in the case of autism, once you know that an individual is susceptible to a particular accumulation in this case or disease state or something, then you can make advice.  I assume that goes without saying to the FDA, that they would take into account that information, that new information on our population.

    DR. APOSHIAN:  That isn't the experience that some of us have had with the FDA during the last 20 years.

    DR. HEIMBURGER(?):  But we have had a change, also, in the FDA now.

    DR. APOSHIAN:  Good for you.

    DR. ACHESON:  May I answer that question, Dr. Miller?

    DR. MILLER:  Of course.

    DR. ACHESON:  Certainly, as new science is published in relation to methylmercury, genomics, proteomics, unquestionably, we would take that into account.  I think the PKU model is a great one, but that is essentially a one-gene issue.  You can diagnose it.  You know early.  You have got the therapy.

    If we could find the methylmercury gene, believe me, I would love to know what it is.

    DR. MILLER:  Dr. Krinsky?

    DR. KRINSKY:  Norman Krinsky.  I am concerned about the consumer response to an advisory that suggests limitation of a certain food type and how that would be interpreted by the public, whether, in fact, as we have heard earlier, that a class of individuals have totally given up tuna consumption, whether, in fact, people might not totally give up fish consumption because of an advisory recommending limitation.

    What comes to my mind is fat consumption in this country.  Despite the commercial success of the Adkins Diet, there is still a huge proportion of the population that believes that fat is bad for you and, therefore, are seeking out non-fat foods, fat-free foods.  The impact of this is that it limits their Vitamin E consumption.

    If they take a multivitamin that contains the RDA for Vitamin E, they are going to be okay.  But that doesn't necessarily go hand-in-hand that people limit their fat consumption and try to replace a missing nutrient.

    There are good things in fish.  I don't know how to address this but I do have a concern that the public might, in fact, move towards eliminating fish as a food source.

    DR. MILLER:  We are going to discuss this at much greater length tomorrow when it comes up.

    DR. KRINSKY:  Okay.

    DR. MILLER:  Jean?

    MS. HALLORAN:  I have a question for Mr. Carrington.  I am just trying to make sure I have

understood your analysis.  So let me read this back to you in another form and you can tell me if I have got it right.

    If you define success as having less than 2 percent of the population with greater than the RFD in their blood of mercury, you would have to have them eating--limiting their consumption either to 6 ounces a week or to 12 ounces a week with the low fish consumption group?

    DR. CARRINGTON:  Of the scenarios I presented, I think that is right.  You could probably devise other scenarios that would--or some combinations of those.

    DR. MILLER:  Anybody have anything else?  If not, then we come to the end of today's work.  We meet again tomorrow morning at 9 o'clock.

    (Whereupon, at 4:00 p.m., the meeting

recessed, to reconvene at 9:00 a.m., Thursday,

December 11, 2003.)