U. S. Department of Health and Human Services
U. S. Food and Drug Administration
Center for Food Safety and Applied Nutrition
Office of Plant & Dairy Foods & Beverages
September 30, 2002

P R O C E E D I N G S

MR. LEVITT: Good morning. Let me welcome you. My name is Joe Levitt. I'm Director of the Center for Food Safety and Applied Nutrition here at the Food and Drug Administration, Department of Health and Human Services, and I'm happy to be your host today, at least during the first session, and to welcome you to this Harvey W. Wiley Federal Building. We're very pleased to be here. We've been here less than a year, but we hope to be here many years in the future. Many of you will become accustomed to coming out here and having public meetings such as today.

As everybody knows, the subject of today's meeting is acrylamide in foods. When you think back just six months ago, none of us were thinking very much about even the possibility of acrylamide in foods. But with the initial findings reported by the Swedish Government, a lot of activity has followed up them, and what we're about to show today and describe today are both the activities that have happened to date but, more importantly, what is our draft action plan for addressing this issue in a very serious way over a short period of time in the future.

In order to do this, we will be using the same tools and skills that we have on many other issues. We'll be focusing on what does the science say. We will be doing this in an open and collaborative way, and we hope very much that will get us to the end game that we need to get to.

We'll be focusing on sound science, on risk-management decision making, but our overall goal is going to be to prevent or reduce the potential risk of acrylamide in foods to the greatest extent feasible. We understand this is a problem, and we want to address it very forcefully and very seriously.

Now, this meeting today has some very simple and straightforward objectives.

Number one, we want to present publicly a scientific overview of acrylamide in foods. Right now, what do we know? What do we not know yet? What do we need to know more of?

Number two is we will present, as we have made public through the Federal Register, an outline in some considerable detail of our draft action plan for how we're addressing this, and we'll have a whole series of panelists in a short time that will be doing that.

And, third, we will be obtaining public comment, both today at this meeting as well as asking for it in the future. We have asked for public comment in just 30 days. That's not the usual time that FDA gives. We usually give 60 days, 90 days, or even more. But I think that underscores the timetable we feel we're on. We feel this is an important issue. We want to move ahead quickly, and we ask your cooperation in our ability to do that.

We'll be going through a number of things. As I said, we'll be starting with what do we know, what don't we know. We'll be talking about toxicology. We'll be talking about how it's formed in foods, where it occurs in foods, and exposure we get from foods.

We will be presenting the first set of findings from testing that FDA has conducted, and we'll talk about additional research plans that are underway. You'll be hearing about interagency efforts as we work with all of our sister agencies, and we'll be talking about the need to develop consumer messages so consumers know what eating patterns they should be adopting.

As I mentioned, the public input is a very important part of our action plan development, and as you know, since I've been director, we have tried to incorporate a strong public involvement in many of the issues that we've been addressing. We want your recommendations on changes to our overall plan, and as I said, by October 30th, which we know is just a month away. We will be moving ahead, as you'll be hearing more later on, to our Advisory Committee process. We'll be having a meeting of a subcommittee in December and our full committee in March. So, again, you see the timeline we're on shows the urgency that we feel on this issue.

Finally, let me just take a moment, before I get the honor of introducing Dr. Crawford, to just introduce the FDA staff that you will be hearing from later today. First of all, Dr. Terry Troxell, who is Director of our Office of Plant and Dairy Foods and Beverages, what we fondly refer to as "land foods"; Dr. Rick Canady, who is a scientific expert in that office; and Dr. Lauren Posnick, who will be presenting some of the research findings to date.

We'll also be hearing from Dr. Tom Sinks from CDC who I think is not here yet--oh, yes, he is. There he is. Hi. Welcome. And, finally, I also want to acknowledge, though he's not one of the speakers, Dr. Steven Musser, who has been very instrumental in developing a laboratory method and heading up the team that is doing the sample analyses here at the FDA.

Finally, I want to thank the organizers from FDA for today's meeting. I find none of these things happen automatically, and so I want to thank Lou Carson and his staff for the effort in putting this together.

So, with that introduction, as they say, let us get on with it. We will have two speakers during this first session. The first--especially since our battery is fully charged.

While we're doing that, there are two other people in the audience that I would just like to acknowledge. You'll be hearing from them after lunch. First is Michael Jacobson, who I see sitting right here, from the Center for Science in the Public Interest, who will be presenting views from the consumer community. And the second is Dr. Dave Lineback, whom I saw right up there, who is Director of JIFSAN, the Joint Institute for Food Safety and Applied Nutrition, our joint enterprise with the University of Maryland, and he'll also be making a presentation this afternoon.

With that, if we can get ourselves up to the next slide, the next introduction, there we are. With that, it is my pleasure to introduce Dr. Lester Crawford, Deputy Commissioner of the Food and Drug Administration.

DR. CRAWFORD: Thanks very much to all of you for being here. As Joe has said, we're going to have some presentations today about what we need to find out and what we know and what we don't know about acrylamide as a potential public health hazard, and it's my pleasure to both welcome you and also assure you that we take this matter very seriously, and we want to hear from all of you and also the general public and informed persons from around the world about what it is that we should be doing. And we also will be working very carefully with international partners in arriving at our plan of action--a plan of action that you will have full access to.

You will also hear how we plan to conduct the necessary research and how we want to utilize its results. We're dealing with a matter of public health significance, and I think we should all recognize that it is.

Acrylamide may present a dietary risk that we believe could be prevented or reduced with the help of the new knowledge that our agency and its partners expect to develop. So the issue before us it not only an intriguing scientific challenge, but it is of primary interest to consumers. That's why we want to outline for you and submit for public comment our plan for characterizing and managing the potential health hazard that might be caused by consumption of food containing acrylamide.

My colleagues will have a lot to say about acrylamide and our action plan, but, in general, this is the issue before us. Last spring, researchers at the Swedish National Food Administration and Stockholm University analyzed food samples, including several favorite American products, and found that they contained various levels of acrylamide. Their study suggested that acrylamide formation is particularly associated with carbohydrate-rich foods that are oven-baked or fried at high temperatures.

Since then, scientists in Norway, the United Kingdom, and Switzerland have come up with similar findings. For example, a meeting at FAO and WHO in June reported that acrylamide levels in 39 samples of potato chips ranged from less than 1.4 micrograms to 100 micrograms per ounce, with a mean level of 15 micrograms. Forty-one samples of soft bread contained acrylamide levels between less than 30 micrograms and 162 micrograms per kilogram, which is a little more than two one-pound loaves, with a mean level of 50 micrograms.

Our own preliminary analysis at FDA appears to be in basic agreement with these results. These are disturbing findings not so much because of what we know about the chemical, which is relatively little, but, rather, because of what we don't know about it.

We know that acrylamide is an organic chemical that is most widely used as a coagulant in water treatment, but it has many other industrial uses, including in the production of organic chemicals, dyes, and plastics. We also know and have known for some time that acrylamide causes or is associated with cancer in laboratory animals. What we don't know is the answers to the very many public health questions that these new findings occasion. The most crucial of these are:

To what extent, if any, acrylamide in food presents a hazard to human health. Epidemiological studies conducted in the past did not show increased cancer risk in people exposed to certain amounts of acrylamide, but the studies were limited in scope and did not include ingestion of the chemical in food.

We, therefore, need to investigate whether acrylamide in food is a potential human carcinogen and whether it is genotoxic, whether it can cause mutations to germ cells. We need to know in which foods acrylamide occurs, what is the level of exposure of the general population, what is the bioavailability of acrylamide in food--in other words, can it be absorbed and in what level given the dosages that are presented for absorption?--and what are the biomarkers of acrylamide exposure.

And we need to know more about acrylamide formation so that if the chemical does prove to pose a risk to humans we can suggest how to modify food-processing techniques to prevent or reduce the formation of that chemical.

This is a big task, and as you will hear in more detail from our colleagues, we're working on it. We're in the process of assessing our consumers' dietary exposure to acrylamide. We're gathering new information about its toxicology, and we're participating in an investigation of how and under what cooking processes acrylamide is formed in food.

We have already developed a method to determine the levels of the chemical in foods, and we are using it to test scores of different products. But as you will hear in greater detail, more has to be done to find scientific answers to the questions about acrylamide.

That would enable us to accurately measure the risk that chemical may pose to public health and, if necessary, try to devise ways of managing it.

Foods rich in carbohydrates are and have been for centuries a robust basis of human diet, and we must make sure that they enhance and not imperil human health. This is the ultimate purpose of our action plan. It's an essential goal for the protection of public health, and I take your presence here as a sign that we can count on your help in our efforts to reach it.

Again, thank you for coming. I hope that you will find our presentations both informative and stimulating as we together strive for solutions.

Thank you very much.

[Applause.]

MR. LEVITT: Thank you very much, Dr. Crawford.

Our next introductory speaker is Dr. Bernard Schwetz, who serves as the agency's senior science advisor. He has been at FDA for a long time, but most recently he served as the leader of the U.S. delegation to the WHO Consultation on Acrylamide, and so he's had firsthand experience dealing with this on a global level, and, Bern, if we could just welcome you up here.

[Applause.]

DR. SCHWETZ: Thank you very much, Joe. And good morning and welcome to all of you. Glad to have you here for this sharing of information and sharing of questions that you might have. We appreciate your interest and the fact that you're here today.

As Dr. Crawford mentioned, the Swedish researchers earlier this year reported information about the presence of acrylamide in a range of foods, particularly carbohydrate-rich foods that were cooked at high temperatures, and subsequently their results have been published in peer-reviewed technical journals.

The finding was certainly of concern to all of us, not because acrylamide is something that we've never dealt with before--it's a chemical that's been around that toxicologists and regulators and people in industry have been dealing with for many decades--but this was a new level of concern for this old chemical because now it was reported to be present at easily measured levels in food.

The concern, of course, comes from the fact that acrylamide does cause cancer in laboratory animals. It isn't known to cause cancer in humans, but it is ranked by the International Agency for Research on Cancer as a probable human carcinogen because of the mechanism by which it causes cancer in animals and the strength of the data in animals. So that clearly made this a chemical of concern and a mode of exposure of concern.

The Swedish scientists were prompted by the observation that they measured hemoglobin adducts--this is a combination of acrylamide and its metabolite to hemoglobin, which isn't an index of toxicity but it's a marker of exposure. There are a lot of chemicals that adduct to various proteins or chemicals in the body, but hemoglobin and its--or acrylamide and its metabolite do cause an adduct to hemoglobin, which is an easily measured marker of exposure.

Well, they found this hemoglobin adduct in more people than they would have expected to find it in and found it at fairly uniform levels and wondered why would this be in so many people who have no known exposure, control kinds of people, people who you wouldn't assume have had exposure to acrylamide through an occupational setting or through a laboratory setting or some other setting.

Well, they did another experiment where they fried rat chow to see if they would create acrylamide by heating rat chow to a high temperature, fed it to rats, and measured--they were able to measure acrylamide in that fried chow. And they were also able to measure acrylamide adducts in the hemoglobin of these rats. So it was another indication that, yes, this was a real signal and that it could be related to food.

Well, subsequently, they developed appropriate analytical methods and developed the methods to the extent that they were comfortable that these were reliable methods in giving useful kinds of information, and they then set out to look at another range of foods to find out how much is there and in what kinds of food, and that's some of the information that Dr. Crawford gave. Certainly there's a lot more information since then, but that was the beginning of the information on acrylamide in food.

I want to point out that it isn't likely that acrylamide is a new component of food, that this is something that's just happened in this past year and we detected it immediately, the Swedes did, and reported it. Because this is probably a function of how food is processed, how food is cooked at home, it's very likely that we've been forming acrylamide in food for a long time. But this is the first time that we have known about it and have known the levels in food. So it isn't a new phenomenon, but it's a new understanding of how much is there.

What's also of concern is the fact that it can be created in food through traditional cooking methods at home as well as the procedures of processing and preparing food commercially.

Well, subsequent to the Swedish findings, WHO and FAO convened an expert consultation on the public health implications of acrylamide in food on June 25-27 this past summer, and three of us from the FDA were participants in that meeting and had an opportunity to join in the discussions as WHO formulated its--WHO and FAO recommended certain actions that should be taken. We were part of that meeting.

Well, as you might expect, acrylamide is a complicated food problem. A lot of things are in food. Just because of the nature of food and the diversity of food and how it's prepared and how it's eaten, how it's stored, there are few things that are simple about food. But acrylamide is one of those things that is also complex in food. And in this consultation, we were brought together to review the information that was relevant to the toxicology, the carcinogenicity, the metabolism, the toxic effects that are known to be associated with acrylamide, either in humans or in animals, to review the epidemiology data that were available, to look at exposure assessment; also to review the analytical methodology that was available because what we anticipated was that this meeting and these findings would stimulate all kinds of work in laboratories throughout the world.

One of the problems that we frequently encounter is that different labs use different methods, different sensitivities, different equipment, different protocols. And we end up with data that don't fit together. So one of the objectives of this meeting was to reach some closure on what are the best methods to use as everybody went back and stimulated more research to find acrylamide in food.

In addition, it isn't just important--it isn't enough to just know how much is in food. The real benefit of knowing what's in food is to figure out how it got there. What are the means by which acrylamide is formed in this process of cooking food or the commercial processes and would give us handles on how to keep it from being formed in food?

In addition, it's a question of the bioavailability. A lot of things are in food that aren't available to the body. They're not absorbed from the food. So one of the major questions here is also how much of acrylamide and forms of acrylamide are available from food. So how much is absorbed? That creates the dose to people.

Beyond these findings of the review of the chemistry and the mechanisms of formation and the toxicology, part of the consultation was to identify needs that involve further data that we should have to be able to make more informed decisions. So what kinds of studies should be done?

In addition, we met and discussed interim advice that countries could use based on how much/how little we knew about acrylamide in food so that we didn't have some countries going off making a totally different kind of statement on their own based on others looking at exactly the same data set. We couldn't keep that from happening, but, nonetheless, part of the discussion was: What are we going to say? What message do we want to get out about the state of knowledge and the state of the risk?

So the consultation, those of us who were there had access to international assessment documents, various kinds of reports on acrylamide. We had background papers that had been prepared specifically for the meeting, and we had new data, we had publications. But we also had some of the world's experts there on toxicology, on chemistry, on a number of other areas so that we had access to those people who had the best information that was available.

Let me give some background kinds of information about acrylamide. As Dr. Crawford already mentioned, there are a lot of industrial uses of this: in grouting materials that are used commercially; there's less in grouting materials today than there had been in the past, but in the past there has been enough in some grouting materials so that in enclosed places, like in tunnels, people using this material were exposed to enough to cause neurotoxicity in humans. Well, grouting materials have been cleaned up, and water doesn't have as much as it used to.

It's also used in water purification. It's also one of those chemicals that's commonly found in laboratories through the use of various gels that have acrylamide in it, and polyacrylamide.

In addition to its presence in food and the occupational settings and commercial products, we also know that there are exposures from cigarette smoking. So that's another source. So we have occupational exposures, we have exposure from cigarettes--that, again, covers a lot of people, but the exposure that you would have there probably isn't as widespread as you would get from food. I'm not commenting on the levels that there are in food, but just the number of people who might be exposed.

Another thing to remember is that acrylamide isn't alone in our concern. Acrylamide has a metabolite, a product that's formed in the metabolism of acrylamide in the body, glycidamide. Glycidamide is also highly toxic. It also forms adducts to hemoglobin. Both acrylamide and glycidamide form adducts to DNA. That's what we're worried about. The attachment to hemoglobin gives us a marker, but the attachment to DNA is what causes the concern.

What about the formation or occurrence of acrylamide? We know that it's in foods that are cooked or processed at high temperatures. Carbohydrates, proteins, amino acids, lipids, natural components of food can be involved in the formation of acrylamide. If you read the article in the Post this morning, you saw a summary of acrylamide in food, and you also saw that asparagine is an amino acid that researchers have found that, together with glucose, might account for the formation of some of the acrylamide in food. So if you have foods that are high in asparagine and glucose, and you have the right temperatures, the right cooking conditions, that could contribute to it. It isn't the given that that's the only amino acid or the only component, but that could be one of them.

It's important for us to have that kind of information to know how to modify foods so that we don't have the conditions that are in favor of forming it, forming acrylamide.

One of the things that we find in most circumstances like this one where you identify a new concern, a whole bunch of researchers go out to investigate what is happening, what you find out is it isn't near as simple as one cause, one chemical, one formation, one food source. We'll find multiple mechanisms of formation of acrylamide, and as a result, the control of it in food will have to involve all of those mechanisms.

One of the things that's true about its formation in food is that increased cooking time and temperature are important and so is low water content. All of these facilitate the formation of acrylamide.

Acrylamide levels in food that are measured may also reflect not only the formation but the degradation. So it's probably being formed under certain conditions and being destroyed in the food. So what you see is obviously what's there, but this is a process of forming and degrading in food all at the same time.

A little bit about analytical methods. There are methods that are available that have complicated, long names that identify the kind of equipment that is used, for the most part, mass spectrometry, mass spectrometric methods, combined with others. One of the things that we asked repeatedly of the chemists who were in this consultation: Are those methods good? Are they the best that we can have? Are they repeatable? Are they methods that can be used in laboratories throughout the world, or are these methods that would only work in one or two laboratories that have the right kind of equipment and the right kind of people?

Well, the consultation was quite clear that these are good methods, they're rugged, they're transportable to other laboratories. The methods have been well worked out. The levels of sensitivity, the levels of detection, all of these things are reasonable. So the consultation expressed quite a bit of confidence in the analytical methods.

One of the things that we're still looking for, though, is validation of this method between a number of laboratories. Again, we're at the scale-up, part of this. One of the things that would be good to have behind us and that people are working on right now is confirmation that if you distribute a set of known samples, including either no acrylamide or graded levels of acrylamide, and ask the laboratories to analyze them all using their method or the standard protocol and standard method, that we have an idea of the variability between laboratories, between chemists, between analytical people in the laboratories doing this work so that that's another critical phase that we're involved in, but that remains to be completed yet.

There are preliminary inter-laboratory validations that are being run through the U.K.'s central science laboratory. Maybe Dr. Lineback will say a little bit more about that later on. And there are also some limited analyses being done by NFPA. So, again, this is the right time to be doing this kind of thing as opposed to four or five years from now when we're knee-deep in doing risk assessments and people raise questions about the analytical methods.

So part of what I think we want to get across today is that we're trying to proceed in an orderly fashion of getting the things done that need to be done so that when we make decisions about risk, we're basing it on accepted data and accepted methods.

The Food and Drug Administration has developed a method, a method that we feel is the one that we want to use in our own laboratories, and it's been posted on the CFSAN website. So that methodology is available. We're not keeping that back. That's out on the website. Dr. Musser's group has analyzed a couple of hundred samples already based on a plan that they have had to analyze samples to be the basis for the plan that you're going to hear more about today. And to demonstrate the validity of the method, those couple hundred samples have been analyzed repeatedly so that we have an idea of the repeatability and the strength of this method. I'm sure you'll be hearing more about that there the day.

Okay. What about exposure? The consultation prepared an estimate of exposure from acrylamide in food, but recognizing that there weren't very many data points that were available--there were just a few foods where there was information. So the amount of information to estimate an average exposure was quite limited, but based on the information that was available, the estimate would be something in the range of one microgram per kilogram of body weight per day level of exposure to acrylamide coming from food.

Now, there are other exposures to acrylamide as well, as I mentioned, but keep that one microgram per kilogram per day in mind.

When you have something like food, you also worry about vulnerabilities throughout the population, and one of the things that is also true in this case is that children would be exposed to a couple of times higher level of acrylamide from foods because of their food intake per body weight. So that's another consideration that we have to keep in mind.

Okay. What about the toxicology profile? It's pretty clear from animal studies that acrylamide causes harm to the nervous system. It is a neurotoxic agent. It's also genotoxic, causes damage to DNA. It's carcinogenic in several studies. And it causes an effect on fertility, an adverse effect on fertility in laboratory animals. Let me give you just a little bit more of each of these.

The consultation concluded that the amount of acrylamide in food is not expected to have neurotoxic or reproductive effects. Based on our knowledge of the levels of exposure that were associated with neurotoxicity in humans, we're trying to figure out how much of a gap there is there between those associations. But at this point, the consultation concluded that the amount of acrylamide in food would not be expected to cause neurotoxicity in humans.

The no observed adverse effect level in animal studies for the neurotoxic effects is about 500 micrograms per kilogram per day. Remember I said the exposure of humans was about one microgram? The lowest dose level in animal studies that is associated with a neurotoxic effect is about 500. The no observed adverse effect level for the adverse fertility effects in rodents is at about 2,000 micrograms per kilogram per day, a couple thousand times higher than what you would expect in people at this time.

But I would remind you again that the data that we have is not just limited to laboratory animals. Acrylamide is well known as a human neurotoxicant, has been primarily from the occupational setting. That's information that's been known for a long time. The standards have been set so that you don't see neurotoxicity in the workplace today, but that was a large part of our knowledge that taught us that acrylamide clearly is a neurotoxicant in humans.

Let me say a little bit more about the adduct information because this is something that can be extremely helpful and valuable to us, but it's also information that can be misinterpreted and used incorrectly.

I talked about the fact that both acrylamide and its major metabolite, glycidamide, cause adducts to various proteins and other components either in food or in the body. Hemoglobin is one of those where there is an adduct that's formed so that acrylamide or glycidamide attaches to hemoglobin, and when you measure hemoglobin, you can measure the amount of acrylamide or glycidamide that is in the blood.

That doesn't reflect toxicity because that particular acrylamide molecule at that point is bound to something that keeps it from getting to the nervous system. So it's a marker of exposure. It isn't a marker of toxicity.

Red cells have an average life of about 90 days to 100 days, sometimes a little bit longer, and then the hemoglobin is recycled. And at that point, what you're measuring for adducts in the blood is an average of exposure over every day. So if you take one blood sample and measure adducts, you're getting an integrated measure of all of the exposures that have taken place in the last 90 to 100, 110 days. So, again, if you had a major exposure at one point in time, you would have a different profile perhaps than if you have had a very low background level every day.

What I'd also mention, though, to remind you, is that it's the fact that acrylamide and glycidamide attach to DNA. That's what is of concern to us, and it is the likelihood that the carcinogenic potential of these compounds are associated with the fact that it binds to DNA and causes a disruption of DNA function.

Acrylamide and glycidamide are also genotoxic. They cause damage to the genes by virtue of their binding to DNA. One of the concerns that we have also is that there's a fair amount of literature on acrylamide in causing germ cell damage, damage to the DNA in germ cells. And the fact that you would have an animal carcinogen that causes some level of germ cell damage in food gives us concern.

Generally, though, the germ cell damage has been seen--we don't know about this in humans. We know about this from animal studies and studies in cells, and for the most part, the germ cell damage has been seen at very high levels of exposure, by routes of exposure that people don't experience. It wasn't orally or it wasn't applied to the skin or it wasn't by inhalation. These were large amounts of acrylamide that were injected intraperitoneally or through some other route of unusual exposure to find out if acrylamide had the potential to cause germ cell damage, not to assess risk. So one of the things that we need to do is continue to find out how do we use those data to predict risk for humans.

I remind you that IARC, the International Agency for Research on Cancer, classified acrylamide as probably carcinogenic to humans, not based on human data. We do have people--we obviously have people all over the world who have markers of exposure to acrylamide, and one of the things that we need to do is if there are occupational cohorts of people who have a fairly high level of adducts, that's a good population to be looking for cancer measures. So I'm assuming that that's one of the things that will be happening if it isn't already.

The epidemiology studies, the people who have been exposed to acrylamide occupationally don't show an increased incidence of cancer, but epidemiology studies are relatively blunt tools, so it takes multiple measures, multiple studies, even if you had a positive, you don't tend to make a decision based on one weak signal in humans from an epidemiology study. You like to see it repeated. At this point we don't have any signals in humans about cancer, but those are the kinds of studies that should be done.

The cancer risk from cooking foods? Well, one of the things that we have come to realize is that there are a lot of chemicals in food. There are quite a few chemicals in food that are carcinogenic in laboratory animals. Polycyclic aromatic hydrocarbons are present in food. That's not acrylamide. Heterocyclic aromatic amines are present in food; they have carcinogenic activity in laboratory animals. There are some natural components of uncooked food that are carcinogenic under certain conditions. So the fact that there is a chemical that has carcinogenic activity in laboratory animals and food, again, isn't a new and novel finding. It gets put on the list of a fairly large number of chemicals that are in food.

Obviously, at low levels, ones that we don't associate with a carcinogenic effect, unless you're doing something to food to really raise the level of some carcinogen by virtue of how you burn it on the grill or whatever you do with it could cause mutagens to be formed in food, but that's not the situation that we expect here with acrylamide.

One thing that's true, though, as I mentioned, a lot of these other chemicals that are carcinogenic in animals that are in food, they're at very low concentrations. One of the concerns we have is that when you're talking about the levels of acrylamide in food that you will be hearing about today, they're kind of high levels compared to some of these other chemicals that are present, but at very low levels. So that's part of what we have to take into account.

The consultation also addressed the question, you know, what should we do about this. One of the general conclusions was that the exposure for such chemicals, carcinogens in food, the exposure should generally be as low as we can achieve. So that clearly would be a goal of ours to not tolerate having an animal carcinogen in food that doesn't have to be there. So part of the research is trying to find out not only how much is there, but how do we prevent it from being there.

The consultation was asked to recommend a temporary provisional, tolerable daily intake, and we chose not to do that. First of all, we didn't have all the people there that would have helped to do that. We didn't have enough data, we felt, and we didn't have enough time in two or three days to be able to come up with a number that would have that kind of importance. So we chose not to go that route and try to come up with a tolerable daily intake or a reference dose.

There are some things that we need, some additional data that we need to have in improving our ability to make this risk assessment and judge what that concern might be for people: further information about the metabolism of acrylamide, how it breaks down across more species, including humans, more information about the binding to DNA and the binding to other proteins. For example, one of the things that is of concern in sperm is the binding to protamine. We don't have enough information to find out the association between binding to some of these critical proteins in sperm and the sperm-damaging effect.

We need to have more information about the bioavailability. When you have acrylamide in food, how much of it can be absorbed by various people, various states of health, various ages. So what is the bioavailability of acrylamide from food?

What are the non-food sources of exposure and how high are they? And are there other sources of exposure that we should be looking at to reduce?

One of the things that was interesting and a little bit of a surprise to me--I just hadn't thought of acrylamide being formed by the body, but the Swedes feel that--and they're developing data. They're asking the question: Could acrylamide be formed endogenously in the body? I'm not sure why the body would form acrylamide, but maybe some of our other colleagues here can answer that question. But it wouldn't be a surprise if we found a small amount being formed in the body as well.

Again, information about the cancer epidemiology, further studies--we haven't done studies in detail on glycidamide, the toxic metabolite, the genotoxic metabolite of acrylamide. We need to have more data on that because sooner or later we're going to come down to the situation of not wanting to just lump acrylamide and glycidamide together as the same decision. We'll want data on both of them.

In addition, while we have a fair amount of information on the damage to germ cells, we need to have more information to figure out what does this tell us to look for in humans and how do we identify what a safe level of exposure to acrylamide might be based on that germ cell damage.

More information on formation, systematic examination of the processing conditions that might contribute to this, and some models that might be derived once we have some information to be able to put models together so that we don't have to continue to measure every conceivable batch of food to find out if it has acrylamide in it and how much, but that we can model what are the conditions of food preparation either at home or in the plant that contribute to this.

In addition, as we begin to have information from these preliminary studies of how much acrylamide is in what foods, that will help tell us how far through the grocery store do we have to go to continue to take samples, because that obviously could be an endless process. But as we learn more, it will give us guidance on what foods, what conditions. So that's a learning process as we go.

From the standpoint of analytical methods, the inter-laboratory validation is something that is important to bring to closure so that we can get standard methods, standard reagents, standard conditions for doing the analyses out there so that when we see information coming out in the literature, we don't have to wonder, well, was this method one that we would accept. We would have better information than that.

From the standpoint of human exposure, the foods from different regions, different diets, there are groups of people all over the world who have various dietary conditions that you don't see in many other places in the world. To what extent are they contributing to acrylamide and to their conditions? And as we move more food across the world today, this becomes a greater concern for all of us because of the importation and exportation of food.

Also in the consultation, the recommendation came out that there should be an international network for sharing data on acrylamide in food, and one of the conclusions that--one of the follow-ups that has happened since then that involves us is that the Joint Institute for Food Safety and Applied Nutrition has volunteered to be that repository of information. JIFSAN is a joint effort between the University of Maryland the FDA, so Dr. Lineback, as the director of JIFSAN will be talking, I'm sure, more about that effort of JIFSAN serving as the repository for all the new analytical information, the new protocols, the new--whatever the methods might be for analysis that that information would be available in one source and it would be through JIFSAN.

So as you see, we have a lot of things going on and, fortunately, we're at a stage where we can still kind of keep track of what's going on in the world and we can be sure that the work that's done will use good methods so that we don't have a lot of misinformation out there or information that can't be trusted.

The consultation also made some recommendations on what people should do about eating food in the meantime, because this does cause anxiety, does cause concern. One of the things was that food should not be cooked excessively because we know that excessive cooking can lead to the formation of more acrylamide. But there's a trade-off there. If food isn't cooked enough, there are biological risks, pathological risks. So, again, it isn't a matter of not cooking food because a lot of the food cooking has to do with making it safe to eat. So there's a trade-off there.

But generally, and, again, as Rick was quoted on in the paper this morning in the article in the Post, the general advice on healthy eating should be followed, to eat a balanced diet and eat a varied diet, plenty of fruits, vegetables, moderate consumption of fried and fatty foods. That isn't just because of acrylamide. That's general health information.

So it's fortunate that what is advisable to minimize the exposure to acrylamide is also consistent with good practices for eating healthy food.

Let me just close by reminding you again that this presence of acrylamide in food isn't something that just happened in the last couple of months. It's been there for some time. And that doesn't make it good or bad. It's just that this isn't something that's totally new about food, but the fact that it's there at measurable levels is new to us.

The concern that drives all of this is the possibility that because it is an animal carcinogen and genotoxic, the possibility that there could be a risk for humans. We are not engaged in these predictions of how many people must die from acrylamide in food every year. The data don't support that kind of a calculation at this point. We don't know enough about it to be able to make that prediction.

In addition, we're planning research, we're conducting research. We're engaged in a large number of studies, everyone from chemistry to animal studies to being aware of human studies as they are planned. Part of the rest of that research plan and the other components of the plan is what you're going to be hearing about through the rest of today.

So I'll leave you with that. Again, thank you for being here. I'm sure that there are questions that you have that will be answerable as we go through the other presentations. So thank you very much.

[Applause.]

MR. LEVITT: That brings us to the closing of the first segment of the program. Let me just quickly outline--you have agendas. Following a short break, which we're about to take but not quite, we'll have a panel of experts that will be going through the action plan and various elements related to that. We'll then have a lunch, and we'll come back for public comment this afternoon.

During the break, you can go out the way you came in. As you know, there are some refreshments out here. The restroom facilities are just a little bit this way towards me down the hall. But I would ask people to come back in about 15 minutes, when Dr. Troxell then will be up here at the podium as the moderator for the next session, and this afternoon, Janice Oliver, our Deputy Center Director, whom I forgot to introduce before, will be moderating this afternoon's session.

So, with that, I have almost 10 o'clock. Let us reconvene at 10:15. That's a few minutes ahead of the written agenda, so it's good to be ahead of schedule. It also means there might be lunch on time as an incentive.

So, again, thank you to Dr. Crawford and Dr. Schwetz for getting us off to a good start today.

[Recess.]

DR. TROXELL: I would like to get the mid-morning session going. I'm Terry Troxell, Director of the Office of Plant and Dairy Foods and Beverages, known as "land foods" and some other names. I won't tell you what my children call it.

Anyway, in this session, I will briefly tell you what we want to cover. First of all, we're going to present our draft action plan. We're going to summarize the interagency research meeting and look how we plan to use biomarkers in somewhat more depth. We will summarize our data from our exploratory survey, and we'll discuss our present consumer message.

First, I'd like to turn attention to introduce the action plan. In developing the action plan, we had several considerations: first, as Dr. Schwetz outlined the science on acrylamide, and quite thoroughly, that science indicates a major concern as exposures are in the tens of micrograms level.

Another consideration that we took into account is also from the science outline, and that is that there's a great deal that is not known.

Third, the research indicates that acrylamide is formed through traditional cooking practices. On the other hand, as you all know, FDA has always been concerned that foods should be adequately cooked to protect the public health from foodborne pathogens.

Another factor is the intense interest stimulated by this problem, as intense as I can remember in connection with chemistry and issues involving foods. Governments around the world are working diligently on it, so are universities and the food-processing industry. Ideally, we would want to bring the power of all these resources together in a coordinated way to address this problem as quickly and as completely as possible to protect consumer health. We tried to capture these and other considerations in developing our action plan.

That brings me to our overall goal, which is, through scientific investigation and risk management decision making, prevent and/or reduce potential risk of acrylamide in foods to the greatest extent feasible. We believe this goal reflects the need for developing the science to inform our risk management decisions, and the goal, through discoveries and risk management actions, to prevent or reduce exposure to the greatest extent feasible.

The plan has several sub-goals. First it is a sub-goal to develop rapid screening methods and validate confirmatory methods of analysis. Methodology is pivotal. One of my general rules is that methodology drives the frontiers of science. It is absolutely essential to have analytical methods, whether for analysis of foods or for biomarkers, to understand the incidence and occurrence of acrylamide in foods, to evaluate mechanisms of formation, and to assess its health impact. For food analysis, we'll be furthering our work on the determinative LC/MS/MS method that Dr. Musser's group has developed in CFSAN.

We also believe it is important to develop screening methods that are rapid and cheap, to support research on formation, and to allow processors and governments to easily determine levels in products.

The next sub-goal is to assess the dietary exposure of U.S. consumers to acrylamide by measuring acrylamide in various foods. This has a couple of dimensions. First, we want to explore more extensively the distribution of levels of occurrence within individual food categories. This will allow us to assess exposures with more accuracy for each food category. The variations we find will also provide suggestions to the food chemists on mechanism of formation and how to reduce the levels.

We also want to explore the food supply more broadly so we understand what I like to refer to as the boundaries of this problem.

The next sub-goal is to identify mechanisms responsible for the formation of acrylamide in foods and identify means to reduce acrylamide exposure. Scientists around the world in university, government, and food industry labs have been working on this for months. That work may already be paying off in the reports that link asparagine, an amino acid building block of proteins, to formation of acrylamide. We must keep in mind, however, that there may be other, perhaps multiple pathways by which acrylamide is formed considering the number of different types of food involved and the complex composition of foods.

However, I am optimistic that chemists--perhaps because I'm a chemist at heart--will find ways to eliminate much of the acrylamide. The challenge may be in taking the findings from the laboratory bench into the processing plant and into consumers' homes.

The next sub-goal is to assess the potential risks associated with acrylamide by evaluation of the available information and expanding the research into acrylamide toxicology, including exposure assessment, bioavailability, biomarkers, and bioassays. This work includes relatively short-term research, which may start paying off in the order of one year, to lifetime bioassays in rodents that can take many years to complete.

The next sub-goal is to develop and foster public and private partnerships to perform successful research. This goal comes from the intense interest by many stakeholders in doing the research. We heard earlier about the WHO/FAO consultation. In that consultation, it was recommended that there be international coordination of the research. The Joint Institute of Food Safety and Applied Nutrition, JIFSAN, our consortium with the University of Maryland, has offered to serve as an international clearinghouse for this research. Dr. David Lineback, the Director of JIFSAN, will be making a presentation this afternoon and will have more to say about this.

We, of course, are involved in many intergovernmental activities, including the Codex Committee on Food Additives and Contaminants and the Joint Expert Committee on Food Additives and Contaminants, both of which will be addressing acrylamide.

JIFSAN and the National Center for Food Safety and Technology, our consortium with the Illinois Institute of Technology, are sponsoring a meeting at the end of October for invited experts from around the world to determine the status of the research and determine the additional research that needs to be done.

We believe such coordination and collaboration by all parties on developing the science base will help us solve this problem most rapidly.

The last sub-goal is to inform and educate consumers and processors about the potential risks and provide options on how to reduce the risks as knowledge is gained. We want to make sure that processors have the latest information on what works and what doesn't work for reducing levels and, for example, that they have cheap screening methods so they can know what levels are in the foods they produce.

For consumers, we would want to be able to tell them how to reduce risks without inadvertently having a significant portion of them more exposed to foodborne pathogens.

I want to take a couple minutes to cover the timetable. Of course, the problem was first brought to our attention by the Swedish Government on April 24th. By June 20th, Dr. Musser and his colleagues developed an excellent method that will be discussed more later. The WHO consultation was held June 25-27, to which FDA had three experts participating.

I guess I'm a little ahead of myself here.

Last Tuesday, we held a federal interagency research meeting which Dr. Canady will also describe more of later. Today, of course, is our public meeting, and I want to urge those who wish to comment to do so by October, if not earlier. I know that's asking a lot; however, if comments are available before October 30th, they may serve as valuable input to the JIFSAN/NCFST meeting.

We also want time to assimilate your comments prior to our Food Advisory Subcommittee meeting in December.

And then, as I pointed out, there will be a JIFSAN/NCFST workshop at the end of October. We will hold the first of two Advisory Committee meetings in early December. We will have the meeting of our Subcommittee on Contaminants and Natural Toxicants of the Food Advisory Committee. Then in March, we will have a meeting of the full Advisory Committee.

Finally, among other future meetings, acrylamide certainly will be taken as new work by the Codex Committee on Food Additives and Contaminants in March 2003. The U.S. delegation will certainly be very active on this matter. Also, JECFA, the Joint Expert Committee on Food Additives, which committee provides risk assessment advice to the Codex Committee on Food Additives and Contaminants, JECFA is tentatively planning a meeting in winter or spring 2004.

Next, Dr. Rick Canady of CFSAN's Division of Risk Analysis will discuss in more detail the major components of the action plan. After that, he will take a deep breath and then report briefly on the interagency meeting that was held here last Tuesday.

Rick?

DR. CANADY: Thanks, Terry. Let me take a minute and pull my slide up.

I want to say that it's really great to see such a breadth of people here today. This is an important issue for FDA, and we need input from all sources.

It's my task today to go through the major components of FDA's Action Plan for Acrylamide. I want to spend a little bit of time while I'm going through the major components to give some of the thinking behind the development of the plan.

When FDA heard of the discovery of acrylamide in foods through the Swedish--through the press releases originating through Sweden, the first thing we did was we did our own preliminary assessment of the toxicology and exposure information. Doing that assessment, we used the information provided by the Swedish National Food Agency and the Stockholm University with regard to occurrence information, where acrylamide occurred in foods. We also used consumption rates that had been developed for U.S. consumers through the USDA food surveys. And then we also used prior dose response information that FDA had generated through prior interactions or, rather, prior decisions that were made with regard to acrylamide.

Using that preliminary assessment, FDA came to three--actually, came to the conclusion that further information was needed, further actions were needed with regard to acrylamide. Defining those actions is what the Action Plan for Acrylamide is all about, determining what actions are needed and how to go about taking those actions.

There were two issues within the assessment that became clear, this initial assessment that FDA did. One is that our confidence in the analytical data that provided information about the occurrence of acrylamide in foods was sufficient for the decision that we needed to take further action. And, secondly, our dose response information, the toxicology about acrylamide, was, again, sufficient for the decision that further action was needed. So we had this initial decision point, this initial preliminary assessment, and it led us to the conclusion further action was needed.

It raised a few questions, too. One of the first questions, obviously, was: Are U.S. foods affected similarly to those demonstrated through the Swedish research? Of course, the Swedish researchers did look at a few foods that were made by U.S. producers, but we needed to have information as to how broad the exposures were, rather, how broad the occurrences were within the U.S. food supply, and to generally confirm whether or not U.S. foods were affected.

We also needed to understand--or one of the questions that came up, rather, was just how is acrylamide getting into food. What causes it to be formed or what causes it to occur in foods?

And, finally--and perhaps this is the more important question with regard to the debate about what to do about acrylamide--we came to the question of: Is there enough information about risks and are those risks sufficient to warrant changing food-processing techniques or to warrant changing dietary practices to avoid exposure to acrylamide?

It's important to keep in mind, as Dr. Schwetz and Dr. Crawford pointed out in their introductory remarks, that acrylamide has not suddenly appeared in food. Acrylamide appears to be a result of cooking, and as it is probably, again, a result of cooking, it's probably been with us for thousands of years. What is new is that we have detected acrylamide in foods, and as Dr. Schwetz pointed out, we were led to the exploration of foods for acrylamide because of biomarkers in people for epidemiology studies.

And, secondly, one of the things that is new about acrylamide is that we've come to the conclusion that cooking does lead to the formation of acrylamide. The details of that are part of the explorations that FDA needs to go through and will be detailed a little more as I go through major parts of the plan.

I've got five major categories of the rest of this presentation, just to help you orient yourself through the talk. I'm going to talk about testing the foods, and what I mean by that is analytical methodologies with regard to telling us where acrylamide is and how much acrylamide is there; and then also testing with regard to telling us just what foods it occurs in, and what that occurrence within food tells us about the exposures that we receive through food to acrylamide.

I want to spend a little bit of time talking about acrylamide formation research, what are the major efforts that we have in order to figure out how acrylamide occurs and what we might be able to do about it.

Then I want to go into the toxicology information, the state of knowledge, what research we're considering with regard to understanding better the toxicology for acrylamide.

I want to move into education, an important aspect of the Action Plan for Acrylamide, helping people understand the risks for acrylamide, helping people understand how it gets into food, what they can do in order to avoid acrylamide, also helping the industry understand what they might be able to do to reduce formation of acrylamide.

Finally, I want to talk about meetings and collaborations that FDA is engaging in in order to expand our understanding of acrylamide formation, acrylamide toxicology, and so on.

I want to spend a little bit of time talking about general issues with regard to testing acrylamide--or testing foods for the presence of acrylamide. One of the first questions, one of the first issues that comes up is: Are we using an appropriate methodology to test for acrylamide? Is it accurate enough? Does it tell us how much is in the foods? Is it precise enough? Does it give repeatability? When you go back to the same food five, six, seven times, does it give you the same answer within acceptable ranges?

A second question is: What is the goal of the sampling? Are we primarily interested in exposure estimation? Or are we primarily interested in formation information? Both of these are important aspects of evaluating acrylamide exposures through food and what we might do about it.

We also need to spend a bit of time--and this is one thing that Dr. Schwetz reflected on in his introductory remarks--about where to start the sampling and where to end the sampling. We need to understand how to go about sampling this newly discovered acrylamide formation in foods.

We know something about where it is through what the Swedish researchers showed us. We know a little bit less about where it isn't.

What foods contribute most to exposure? That's another issue that comes up. If we understand that concentrations vary widely across foods, that's an important piece of information. But perhaps more important, we need to understand which foods contribute most to exposures.

There are some foods that perhaps are considered--that have lower concentrations than others, but we eat them at higher degrees. So we need to consider the information about exposure along with the information about occurrence.

Finally, I have several priority statements up here that I wanted to provide to you and have you reflect on. First of all, is our priority exposure assessment leading to regulatory measures? Should that be our primary focus at this point? Or should we be looking at occurrence variation with the goal of reducing exposures through reducing the amounts in foods?

And stepping back even further, maybe our top priority is finding out where the acrylamide is, simply so we're not missing the big picture. Again, we know a lot about where--or we know something, rather, about where it is. But do we know enough about where it isn't so that we're not missing the big picture?

Okay. Moving into the specifics of the plan, CFSAN has developed an analytical methodology for testing acrylamide in foods. This is this long term that Dr. Schwetz referred to, the liquid chromatography/tandem mass spectrometry. I have to look at it still because it's such a difficult thing for me to remember, for me not being an analytical chemist.

We have developed a methodology. We've tested the method's performance relative to other methods used by other laboratories, including this method used by other laboratories, and we found that our method performs very well, and a lot of credit goes to Dr. Steve Musser and his group for developing this methodology.

We're in the process of developing a formal inter-laboratory validation for the method, again, so it can be used, as others have pointed out, in a reproducible way across laboratories and in a way that informs risk management decisions.

CFSAN plans to update the method as needed, and we have, as many of you are aware, posted our method on the Web. We will update the method as necessary to improve use and portability and transportability of the method and accuracy as needed.

CFSAN will also spend effort trying to develop new methodologies that use less specialized technologies, that have the ability to provide results quicker, cheaper, and so on, with the idea of helping spread the understanding of where acrylamide occurs and facilitating our abilities to reduce acrylamide in foods.

Our sampling plan with regard to acrylamide, again, this gets to the question of where to start and where to end. We started with the foods that Sweden identified. This was a logical starting place. We wanted to confirm occurrence as our first step in testing for acrylamide.

Then we asked our food chemists and our processing technology experts to help us understand what other foods we might look at with regard to acrylamide occurrence. So we expanded our testing to include foods that were like the ones tested by acrylamide that might shed further light on the variation of acrylamide across foods.

We need to within this survey of occurrence of acrylamide probe the occurrence, probe where it is and where it isn't across foods. This has a couple of functions. It helps understand exposure, obviously, but it also helps us understand process, mechanism of formation issues, so that when we move forward to taking action with regard to acrylamide we have more information.

We're going to include--or we are including and will expand the inclusion of foods based on consumption rates, obviously, so we're interested in processing and chemistry and so on, but we're also interested in making sure we've touched the bases with regard to how much food is exposed. If a particular food is consumed to a large degree, we want to make sure we look at that food and see whether it does or does not have acrylamide in it.

Next, after we've surveyed the occurrence of acrylamide in foods, we want to look into variability issues, and this turns out to be a rather key aspect of the evaluation of acrylamide exposure and occurrence. We need to understand how it varies across different foods, across different lots of the same foods and so on before we can understand the exposure to acrylamide adequately to take actions and so on.

I want to point out, though, that this exploratory--this need to explore with regard to variability, with regard to occurrence, doesn't necessarily rule out the need to take actions. We are exploring and we are evaluating the occurrence of acrylamide.

We also intend to evaluate the exposure of acrylamide through use of what we call the Total Diet Study, and this is a market basket survey that I'm going to go into in further detail in the next few slides. But before I go into the Total Diet Study, I wanted to give you an idea of the overall level of sampling that we're currently engaged in or are planning.

With regard to occurrence, the survey of acrylamide across foods, we're looking at a level of sampling right now on the order of 600 samples. We will evaluate whether we need to add more to those 600 samples.

To give you an indication of scale, the USDA's survey of intake lists around 6,000 foods. We don't need to examine 6,000 foods because a lot of those foods are repeats. You have different variations of stews, different variations of various breaded products. But we do need to take enough foods--again, not approaching the 6,000, but enough foods so that we understand variations in ingredients, variations in different types of foods.

The 600 samples does not include samples that we're taking with regard to process evaluations, an ongoing aspect of FDA's Action Plan for Acrylamide. We understand through the National Food Processors Association that they're developing sampling for acrylamide in foods, and they have indicated to us that they're going to make that information available to us through the Joint Institute for Food Safety and Applied Nutrition that Dr. Troxell referred to and that Dr. Lineback will be referring to later this day, we understand.

In addition to occurrence information, the level of sampling for exposure is around 1,000 samples for next year, and, again, this is through the Total Diet Study that FDA uses for a variety of contaminants, and we're going to use it to probe the exposure of acrylamide through foods.

I've got two slides here that are going to look very much the same. In the first one, I'm going to talk about testing foods for occurrence and our fiscal year outlook, and the second one, I'm going to go through testing of foods for exposure. So I'm just going to flip back and forth to give you an idea. They're going to look very much the same, and I'm going to be talking about occurrence in one and exposure in the second.

With regard to occurrence, in the fiscal year that ends today, we have confirmed occurrence within U.S. foods. We've used that information to inform processing research and sampling scope for exposure estimates. We're at the level of approximately 150 food samples, and that includes multiple analyses. I think something on the order of 500 or 600 different analyses have been done for foods by CFSAN over the past--what, since early June. We're about a quarter of the way through the 600; in other words, that I mentioned in previous slides with regarding to testing for occurrence.

For fiscal year '03, analysis of foods across the country will be included, again, to get an understanding of occurrence across different food types, and I've already mentioned the National Food Processors Association.

With regard to exposure, in fiscal year '02, again, which is just ending today, we've confirmed the occurrence of acrylamide in foods, and through doing that given a general indication of the level of exposure--that the level of exposure, rather, through U.S. foods, is on a similar order of magnitude to that which was estimated through the WHO/FAO consultation in June. Again, in fiscal year '03, we'll move into the Total Diet Study, and why don't I just go on to those slides and talk about the Total Diet Study.

The Total Diet Study is a broad survey, and FDA's general purpose in using the Total Diet Study is to determine levels of various substances--in this case, acrylamide, obviously--in an average national diet. It involves analysis of foods representing all components of the diet. Foods are prepared as eaten--in other words, we buy the foods, prepare them as you would eat them at the table. And then we will analyze them for acrylamide.

Intake is estimated using national food intake surveys. Focus is on the average diet, again, but it is subdivided for gender and age categories. And it's part of an ongoing, well-established dietary exposure program for contaminants, including, for example, lead, mercury, cadmium, arsenic, other contaminants that we've evaluated through this mechanism and used to help inform our decisions.

The collection of Total Diet Study samples occurs across the country using four market baskets, as we call them, throughout the year. Each market basket is taken from a different region of the country, and that market basket includes foods taken from three separate cities within that region of the country.

I am giving you this slide in order to give you a sense of the broad scope of the analysis and the strength of the program in establishing exposures for the entire nation.

I want to give you a sense of the status of FDA's occurrence data, and I've already talked about each of these points. Again, we have a range similar to that reported in the WHO/FAO consultation in June of this year. We've confirmed that the range is similar to that.

We have initial indication, as other laboratories have provided, that cooking temperature and cooking time do have an effect on acrylamide occurrence. We're still surveying occurrence. We're still in the middle of the process of determining where acrylamide occurs. We will move into variability evaluation next, again, across foods and within foods. And then, again, we'll go into the Total Diet Study. But I wanted to emphasize that we do not have enough information to adequately explore variability for exposure to acrylamide across the population, different sub-groups at this point.

So, to recap, FDA is collecting occurrence data to identify the scope and nature of the issue and to inform process evaluations. And what I mean by that is processing as it causes acrylamide formation in food, whether in industry or whether in the home.

CFSAN will use the Total Diet Study beginning--I guess tomorrow is fiscal year '03, but beginning fiscal year '03 to look into the exposures of U.S. consumers to acrylamide through food. And this exposure information will help us evaluate the risk of acrylamide in food.

Moving on to the second part of the five major components that I'm going to talk about today, acrylamide formation, we are working with the National Center for Food Safety and Technology and are investigating mechanisms of acrylamide formation, processes for reducing acrylamide formation, and as Dr. Troxell pointed out, NCFST is an FDA-academia-industry consortium.

FDA is also working with the Joint Institute for Food Safety and Applied Nutrition to foster data needs discussions and data sharing relevant to formation mechanisms and effects of processing.

FDA intends to interact with industry to conduct research on acrylamide formation and research again through consortium mechanisms and adapt processes that safely reduce acrylamide.

I want to spend a bit of time talking about our state of knowledge with regard to toxicology, and Dr. Schwetz spent a fair amount of his presentation this morning also discussing this.

There are three main points that I want to make with this slide. The first is that there is an existing database with regard to toxicology. This current situation is part of an ongoing process of evaluating the toxicology of acrylamide. Prior decisions have been made through evaluation of that database for uses with water purification and polyacrylamide, for example; uses in cosmetics; uses in food contact materials. Those decisions, as an example, FDA's decision with regard to food contact involved different scenarios with regard to exposure levels and our choices, our risk management decisions.

Our knowledge has changed with regard to how we are exposed to acrylamide and how much we're exposed to acrylamide. In prior decisions, we had--okay. I'm not doing that.

[Laughter.]

DR. CANADY: At least I don't know that I'm doing that. Sorry about that. Maybe if I just step away it won't do it again. That's just to keep you awake, and keep me on my toes, I guess.

Our knowledge has changed with regard to the amount to which we're exposed for acrylamide, and it's also changed with regard to how we're exposed. Prior decisions were made with the assumption that the exposure happened through air pathways, dermal pathways in the occupation settings, through water pathways primarily for water purification uses. We have a new situation where the primary exposure is through food routes.

There are a couple of ways that that affects our decision making. One is, first of all, obviously the levels are different. Second, the route of exposure and the bioavailability is something we haven't evaluated. That's something Dr. Schwetz talked about. And, finally, we also have a lot more people exposed than we had previously considered. So the information with regard to toxicology, we have to look at it in terms of is this enough information to make the decisions we need to make with regard to acrylamide exposures through the food supply.

FDA's National Center for Toxicological Research, NCTR, is undertaking a series of studies, and I want to begin by talking about the short-term studies that NCTR will be undertaking.

First of all, as Dr. Schwetz pointed out, we want to look into the bioavailability of acrylamide in food and compare that to previous information with regard to how much acrylamide you get in your body by drinking water. A lot of the studies--in fact, most of the studies on which our toxicology information is based were done using drinking water exposures to animals. So it's important to understand how that information differs--or how that information is useful in terms of making decisions for exposures through food.

NCTR will also look into mechanistic information with regard to DNA adducts that Dr. Schwetz mentioned earlier. Adducts are, again, as Dr. Schwetz mentioned, reaction products between acrylamide and proteins or DNA. Adducts give us a lot of good information. When they're available they're a very useful tool for understanding exposures and risks. They can tell you how much you're exposed. They can also help provide a tool for comparing toxicology information from animal studies to applications to human exposure situations. So they're very important studies. These studies, again, as Dr. Troxell pointed out, are studies that will provide information to us in the short term and will be useful in developing decisions in the shorter term.

On the longer term, FDA intends to nominate acrylamide and its metabolite glycidamide to the National Toxicology Program for study through that program, which looks at a variety of endpoints, primarily focusing, as most of you, I think, are aware, on carcinogenicity as an endpoint. But there are evaluations of other endpoints that will be available to us throughout the course of that investigation.

We're going to ask for subchronic toxicity studies, obviously chronic carcinogenicity studies, and look into mechanistic studies. NCTR is going to conduct these studies--at least, that's the intention--through interagency agreement with the National Institute for Environmental Health Sciences. And FDA intends to participate with all aspects of developing the experimental protocols to assure that our regulatory needs are met through these studies.

I want to spend a bit of time talking about non-cancer endpoints. Dr. Schwetz mentioned mutagenicity, germ cell mutation, as one of the effects that FDA is concerned with. We're quite interested in this, obviously, and interested in evaluating the information from the toxicology database, from the animal studies and so on, as to how that applies to the human exposures we know about through food now. Also, neurotoxicology. We need to be cognizant of, again, prior uses of the dose response information and how that applies to our current situation of wider exposures and higher levels of exposure.

And I wanted to point out that the U.S. EPA has developed a reference dose for acrylamide that is based on neurotoxicology. So, again, we need to pay attention to that endpoint.

With regard to education--and this is the fourth of the five major components and we're almost done--FDA will develop educational material to inform and educate consumers and processors about the potential risks of acrylamide in food. And as knowledge is gained, we intend to provide options on how to reduce those risks. And throughout the process of developing these messages, as Dr. Troxell will talk about in some more detail later, we intend to engage stakeholders in soliciting comments with regard to how to frame those messages.

Okay, the last part. Meetings and collaborative projects. FDA has sponsored several meetings, including one that occurred just last week that I'll talk about in a little bit more detail in a minute. There was an interagency roundtable where the primary purpose was to develop--or to find out what research was going on and try to coordinate research. You're at the second bullet; this is the public meeting on acrylamide. And in December of 2002, this year, and March of next year, we intend to have further public meetings with regard to our action plan for acrylamide.

I want to point out that the information that's gained through the JIFSAN workshop that Dr. Troxell mentioned, through this public meeting, through the interagency meeting, and other mechanisms, will all be used to update the public plan for acrylamide and will be taken to the Food Advisory Committee for its input.

Meetings and collaborative projects with regard to consortia and symposia. Again, we've talked about the JIFSAN workshop that's coming up at the end of October. One of the outcomes of that workshop will be data needs and responses for basically the topics that were covered at the WHO/FAO consultation. There are also a couple of scientific meetings that are coming up where we intend to provide information and ask for feedback. One is the Emerging Issues in Neurotoxicology Conference that happens in Little Rock, Arkansas, in November, and the second is the Society for Risk Analysis meeting that happens in New Orleans in December of this year.

International meetings and collaborative projects. We have mentioned several times the June WHO/FAO consultation. We will provide information to and obviously use the JIFSAN clearinghouse that was referred to. JECFA meetings have been tentatively scheduled for 2004, and if invited, we will obviously participate in this exercise in order to do safety risk assessment for acrylamide.

And as Terry mentioned, the Codex Committee on Food Additives and Contaminants is meeting in March and will also meet in 2004, and we expect acrylamide to be an issue with regard to these meetings.

The European Union, I just wanted to point out, has developed an inter-laboratory collaborative analysis project, and the results of this collaborative analysis are already available. FDA, again, did quite well in this collaborative project, and it was an opportunity to compare our methodology to other methodologies in other laboratories.

So I wanted to recap, and this is the goal statement that you saw in Dr. Troxell's presentation: Through scientific investigation and risk management decision making, we will prevent and/or reduce potential risk of acrylamide in foods to the greatest extent feasible.

Here's the point where I take a deep breath, get another glass of water, and I'd like to move into the next agenda item I'm down for, and that is the interagency meeting on acrylamide that happened last week, on Tuesday.

It was held here in this very room. The agencies participating are listed here, and you don't need to take notes. You'll have the handout so you don't need to write all these down. But there was wide participation across agencies that have had experience dealing with acrylamide, again, for prior decisions, and also including agencies that have a new interest in acrylamide through the exposures through food.

The topics addressed at this meeting--and these are the questions that were posed to each of the participants of the meeting: What research is needed to improve the risk characterization of acrylamide relevant to food exposures? We asked them to give us their responses to that question. What are the priority data needs? And what sequencing, once we've identified those data needs, is appropriate? And, finally, we wanted to ask them to identify areas of research, of overlap and potential coordination between the agencies, or with outside parties where we knew of those research efforts.

Additional goals of the meeting were to identify and clarify the status of unpublished or ongoing research. We actually found out about research that we were not fully aware of at this meeting, and that's an important part of the information gathering.

We wanted to identify key individuals for collaborations across the agencies to make sure that these efforts happen as seamlessly as possible. And, obviously, we wanted to start collaboration in support of our decision needs for acrylamide. So we wanted to add to and provide synergies to the ongoing research that helps support FDA's decision needs.

Some of the collaborations identified: There are ongoing National Institute of Occupational Safety and Health studies, NIOSH, that were identified, and we talked about collaborations between various centers at CDC and National Institute of Environmental Health Sciences with regard to exposures to acrylamide and some of the effects that might be monitored in those epi studies that NIOSH is doing.

We talked about mechanistic studies development with regard to collaborations on biomarker analysis methods; in other words, making sure that we are pushing as far as we need to push with regard to the sensitivity for biomarker evaluations. Will the biomarker evaluations be able to allow us to look low enough with regard to our decision needs?

We talked a fair amount about the public health implications of the mutation research that is out there and talked about collaborative efforts that might move that issue forward.

Two work groups that we developed--that we talked about developing, rather, within the interagency working group were, one, to look specifically at biomarker analytical methods, and then a second group, an overall work group, because we recognized the need to continue these discussions and look for collaborations and synergies throughout the agencies that have experience with acrylamide.

The next steps: We intend to edit and finalize the responses that we got to our questions, solicit more interaction and buy-in from the agencies through editing of that document, and update the overall action plan as appropriate through that information. And, again, we intend to establish interagency working groups in order to continue and aid in that collaboration.

That's the end of my show for today. What I'd like to do now is introduce Dr. Tom Sinks from the National Center for Environmental Health at CDC in Atlanta, Georgia, and Tom's going to give us information about the National Center for Environmental Health's work with regard to acrylamide biomarkers in the human population.

DR. SINKS: Good morning. Thanks, Rick, for getting that up. I got up at quarter of 5:00 this morning, and unlike Joe Levitt, I am used to getting up early in the morning since I have a two-month-old. But I don't necessarily enjoy getting up at quarter of 5:00 in the morning. It was a nice flight, and this is such a great facility. As a government worker in our facilities in Atlanta, I can just say this is a wonderful place to be, and it's just fantastic that it's right across the street from the Metro. It's extraordinarily convenient.

Before I start, I do want to talk a little bit about the importance of collaboration between CDC and FDA. My own personal experience in this actually goes back to 1991 when I moved from another part of CDC, NIOSH, and came down to the National Center for Environmental Health. And within one month, the first two outbreaks that I was asked to investigate involved products that either FDA does regulate or has decided not to allow into the United States. One of them was milk which was an epidemic of hypervitaminosis D in Massachusetts due to the over-fortification of milk with Vitamin D, which did not turn out to be a national issue. It only turned out to be a local issue. The state FDA was involved. And the other was an epidemic of akee poisoning in Jamaica, the national food of Jamaica, which has a toxin in it which causes Jamaican vomiting illness and a number of deaths. But it immediately--those two epidemics really within the first month of my arrival at CDC really struck the importance to me of how important it is for us to be collaborating with our other agencies and for us to know those agencies and know the people there and who to contact and that activity. And I can tell you that--many of you may not know CDC that well, but we collaborate with FDA in a tremendously wide area of activities, probably most of their centers. Biologic diseases caused by biologic agents in the food supply, we're actively involved with them. Other examples are the safety of drugs with thimerosal in vaccines. The National Pharmaceutical Stockpile, which is in my center, has issues of licensing of drugs to be used by the Pharmaceutical Stockpile during a terrorism event. Methylmercury in fish, I've worked with Rick on that issue. Many issues, even including folic acid as fortifying food with folic acid to prevent neural tube defects, is a very good example, I think of the positive collaboration between FDA and CDC.

And I think that--I'm hopeful that this will be another very good example of that collaboration. We're certainly very eager and interested in helping FDA tackle what I think is an extremely hard science problem.

Today I'm going to really talk to you very briefly about three things: first, I'm going to describe our biomonitoring program which we have in our center; I'm going to then specifically turn to what we're hoping to do with acrylamide; and, finally, I will identify some critical needs that I think we need to include in our thoughts to get this work done. And let me see if I can operate this. All right.

First, just a description of our biomonitoring program and what we mean by biomonitoring.

To us, in our laboratories down in Atlanta, we look at biomonitoring as measuring environmental chemicals or any other type of agent in human blood and urine. In our laboratories, we do not generally do environmental sampling. We really stick to trying to be the best and the most proficient at measuring things in human blood and urine.

We do it for several reasons, and three of the examples here--let me see if I can get this to work--actually put some basis on those reasons. But first and foremost, we try to determine who is really exposed.

As an epidemiologist who knows how blunt, as Bern Schwetz said, how blunt epidemiology is as a science, to me the critical piece of epidemiology in occupational and environmental health is usually the inadequacy of our exposure measurements. We frequently use questionnaires to determine who is exposed and too infrequently do not use more objective measures of exposure, like environmental samplings of air or, in this case, measuring things in people's blood or urine. So biomonitoring is very critical for us to know who is exposed.

The next step is really building the dose response. You have to know the distribution of exposure. How exposed are people? And then you have to be able to use that information to relate it to hazard in some way. It's not just enough to know how exposed somebody is if you can't pin that distribution in some way in terms of hazard. And that's the issue: Does exposure cause disease?

Another issue is: How does exposure vary across populations? One of the things Bern mentioned this morning was--I think I heard him say that the exposures that were measured for acrylamide in human tissue seemed to be higher than expected, but seemed to be also fairly uniform. For biomonitoring to be very relevant, we're not only going to have to know how exposed people are, but we're going to have to see some type of distribution in that exposure to be able to relate it to hazard. If everybody is exposed to the same level, it's just an epidemiologic principle, we'll never be able to associate that with disease. A good example of that would be smoking. If all of us were cigarette smokers, we'd never know that smoking caused lung cancer.

And, finally, perhaps the most relevant information for us on biomonitoring is: Do interventions work? And sometimes we're very fortunate to be able to use our technology, if you will, to really take a look at interventions. One of the interventions we looked at last year was continine levels, which is a metabolite of nicotine, and we looked at continine levels as they changed from the early '90s to 1999 and found a tremendous change in people's exposure to sidestream smoke as measured by continine level. And it really to me indicates a significant public health success story that should end up with lower lung cancer rates in our population.

I'm going to use this as a further example of biomonitoring. One of the best examples, I think, of an environmental health success story in the past two decades has been the shift in the U.S. population's blood lead levels for the entire population of the United States. And another reason to show this data is it demonstrates our long-term partnership with the National Center of Health Statistics, another part of CDC, which runs the NHANES survey. And our laboratory has been the primary laboratory doing a lot of the measurements for NHANES. And we were doing measurements for blood lead back in the mid-1970s when, in fact, the median level for blood lead in the population was at 16, well above what today we recognize as the level of concern of 10 micrograms per deciliter.

The success has been that--first we used the biomonitoring data to help support the EPA's promulgation of taking lead out of gasoline, which occurred about, I think, 1981, and then we were able to track how that regulation and taking lead out of solder, tin cans, and lowering the lead levels in water have really created a tremendous reduction in the U.S. population's blood lead levels. By the time a decade ago we were down to 2.7, and we will be announcing that this number is fairly correct, that the median level is about 2.

We're really talking now about just 1 or 2 percent of the U.S. population being above this level of 10, whereas two decades ago, we were probably up to about 80 percent of children in this country.

Now, we have shifted our biomonitoring program from being somewhat ad hoc, if you will, to more of a programmatic perspective. We are now working consistently year to year with NHANES to provide to the public, our federal partners, and others population distributions of environmental exposures for the U.S. population. We hope that that data will be useful for a wide variety of issues, not only our own but those of our partners like FDA, and we hope that it will be useful for research purposes as well as forming policy and for evaluating how our interventions work.

This program is mandated by Congress. Congress does give us specific funds for our biomonitoring program, beginning in, I think it was fiscal year 2000. And as I said, the current way our biomonitoring program works is to capitalize on the NHANES survey, which I won't go into in much detail here, but all of the strengths and limitations of the NHANES survey are certainly embedded in what we do in biomonitoring.

Last year, we released the first report from that environmental health report. It was limited to 27 chemicals. They included phthalates. One of the findings was that the high-production, high-volume phthalates were actually not the phthalates that we were seeing in high concentrations in people's blood, but it was actually other products that were used more in consumer products. We also released information on methylmercury which was reported demonstrating that about 10 percent of women of reproductive age are at the equivalent level of the reference dose for methylmercury, but that children aged 1 to 5 have much lower levels.

We also released some metabolite information on organophosphate pesticides, and hopefully late this year we will be coming out with our second report, which will have at least three times as many chemicals.

The other thing I want to stress is that acrylamide is not on that list. This has only--as I think was mentioned earlier, it has only come to our attention recently that this is a national issue. It does take us about a year or so to plan for the development of these bioassays with the national survey going on. And so we have to roll it into our thinking about how to incorporate it into the NHANES program.

Okay. On acrylamide specifically, we do not currently have an assay up and running. You've heard that others do have an assay up and running. It is on hemoglobin adducts. It is for both acrylamide and glycidamide as well, and we are developing that assay. For us to develop an assay, we do require a lot of--we want to make sure that we have good standards for it, that it's reproducible, that we can put it through in high throughput, and we can repeat the analysis from year to year. And the estimate from our laboratorians is that we'll be able to roll with this in about a year's time.

We'll also be trying to provide some standards for our labs and other labs for quality assurance and quality control in these efforts. And as was mentioned before, this is a hemoglobin adduct, so it does reflect a cumulative intake, if you will, over about a 90-day period.

This is a simple schematic of the methodology. I'm not a clinical chemist, I'm not a laboratorian, so don't ask me to go through it. It does use liquid chromatography/tandem mass spec that was mentioned before.

Some of the other critical steps that I think are really important here. First and foremost for us is, if we are going to include this in NHANES, we really need to be planning right now to be setting aside blood samples for analysis. I have looked at what blood has been stored in the NHANES '99 through 2001 collection. Actually, it's the 2002 collection, and my understanding is we have inadequate material of whole blood that has been stored for this use. So if we're going to be adding this to NHANES, it will have to be on the ongoing collection, which starts in 2003. They do a two-year sample, and because we don't have the assay up and running, we're going to have to propose to NHANES that they store whole blood for the purpose of doing hemoglobin adducts in the future.

As Rick had mentioned, we will be working with FDA to establish a working group on the biomarkers laboratory issues. Hopefully we might be able to do some kind of round-robin testing with that. I haven't heard from our lab people if we specifically were thinking of doing round robins with some of the labs around the world, but it is something we do with many of our other assays. I'm not aware that we've done it with hemoglobin adducts.

The other side of this is that while we do have the program with NHANES to do population distributions, that's not going to be adequate to simply tell us what the hazard level is related to biomonitoring. We're going to have to see that these types of measurements are also performed in any epidemiologic studies in order to relate what we might find in the population distributions to what we actually know about hazard levels. So looking for populations that have high levels of exposure perhaps occupationally and seeing that they also use similar biomonitoring levels and that we can correlate those levels with the ones we're doing for the population.

The other thing I think we're going to have to do is look at the exposure sources and know what's important. This is another possible opportunity with NHANES, but I can tell you, I looked at the 30-day recall questionnaire in NHANES that exists right now, and I don't think it's adequate for identifying the specific foods that might be at risk, the fried foods or the foods that we're interested in. And if we're going to correlate the population-based biomonitoring data, we probably need to see that the questionnaire data is adequate as soon as we have that information.

I think that's my last slide. Again, I just want to summarize by saying that, as public health agencies, FDA and CDC do work together. We work together frequently. Most of the time we work together well. And I hope this will be another good example of that collaboration.

Thank you.

[Applause.]

DR. TROXELL: Thank you, Dr. Sinks, for that very informative presentation.

While we're getting this cued up, I will point out that the presentations will be available after the lunch break, so you don't need to--all of you who have been scrambling to take notes, they are going to be with you in a of couple minutes.

Thank you also for pointing out the importance of the collaborations. We, too, believe that the synergies between CDC and FDA have really enhanced the public health.

As Dr. Sinks pointed out, the biomonitoring results on lead show a tremendous reduction in lead levels in the blood. Also, our Total Diet Studies, you know, are our weather vane for these kinds of contaminants in our food supply, and they also track a very--a tremendous reduction in exposures, and that's also one of the values of looking at acrylamide with Total Diet. We'll be able to not only measure what the exposures are currently, but we'll also be able to assess the progress.

Now I would like to introduce Dr. Lauren Posnick in our Division of Plant Product Safety, and she'll be discussing the acrylamide testing we've done and the results of our exploratory survey.

DR. POSNICK: Okay. As Terry said, I'm going to talk today about our acrylamide testing at CFSAN, our exploratory survey results.

This is an outline of today's talk. I'm going to cover the goals of our testing program. I'm going to speak briefly about our method for acrylamide analysis. We've already heard somewhat about that, but I'll give a little more detail. I'm going to discuss our overall sampling strategy and remind you what Dr. Canady said, and explain where the exploratory testing part fits in. Then I'll get to some of the results of the testing to date, and I'll finish up with a summary.

The two main goals of our exploratory testing program are to conduct an initial survey of levels of acrylamide in U.S. foods and, maybe more importantly, to identify the types of foods that we need to look at in the future.

The method, as was already mentioned, was developed by Dr. Steven Musser and his colleagues in the Office of Scientific Analysis and Support. It's a liquid chromatography/tandem mass spec method, or LC/MS/MS, and it was posted on our website on June 20th. Dr. Musser's group made some updates on July 23rd. That's available for the public and other scientists to use. If more updates are made, they will also be posted on the website.

The method is quite sensitive. It has a limit of quantitation, or LOQ, of 10 parts per billion, ppb. The validation for the method is ongoing now. A single-lab in-house validation is being completed. Dr. Musser's group has collected a lot of data for that, and they're in the process of writing it up now.

Dr. Musser is also planning several inter-laboratory validations. The first one will be a performance verification which the method is taken to some other labs and we see how well it handles in the other labs' hands. And he is also pursuing an AOAC type inter-laboratory validation.

I'd also like to mention the current results of an international proficiency testing study that's being administered by the Central Science Laboratory in the United Kingdom. In the latest round of this survey, 33 international labs, including the lab here at CFSAN, were sent a sample with an unknown amount of acrylamide. Results were sent back, and the assigned value of the sample's acrylamide was determined from all the accumulated results. FDA's value, FDA's result came within 3.5 percent of this assigned value, and that's a very good result and reflects well on our method.

Okay. Now I'm going to talk about how we selected samples for our exploratory testing, but I'd like to explain that in the context of our overall sampling strategy, so this will be a little bit of a review of what Dr. Canady went through.

Our starting point is to look at foods that were previously identified as being relatively higher in acrylamide, for example, foods identified by the Swedish National Food Administration. And then we looked at comparable foods in the U.S. food supply and asked: Can we confirm those results? What levels do we see for those types of foods in the U.S. supply?

Then we want to, again, as part of our overall strategy, expand that sampling set based on a number of criteria, including having a wide breadth of foods, not just the foods that were tested already; foods that reflect consumption by different groups in our population, different ages, for example.

Mechanistic insights, we've already talked to our food scientists and asked them for suggestions on what foods we can look at that haven't been looked at so far that, based on what they knew at the time or what was known so far about mechanisms of formation, what foods they thought might contain acrylamide. We've already gotten some suggestions from them, and we will certainly get more in the future.

Consumption rates, we want to make sure that we're getting foods that are important, that are high in consumption in the American diet, not just foods that are, you know, potentially high in acrylamide but low in consumption. It's important to get the foods that reflect the major consumption areas of the diet.

Two other parts of the overall sampling strategy: We want to, again, evaluate variability in foods by looking more closely at foods, and also evaluate exposure to the population, taking advantage of the Total Diet Study, which, as you've already heard, reflects the average diet in various age and gender groups in the U.S. population.

Okay. Now, where we are now in this strategy, we're kind of partway through that sampling strategy. We've done some confirmatory studies. We've expanded the breadth of sampling to an extent. And we've made some sampling decisions based on mechanistic insights. That has led us to about more than 150 food products, and for those products, about 500 analyses.

We've divided the foods up into categories. They're shown here: baby foods, canned vegetables--that might be a little bit of a misnomer; it actually is beans and mushrooms--cereals, chocolate products, coffee, cookies, crackers, French fries, gravies and seasonings, infant formulas, a miscellaneous category, nuts and nut butters, potato chips, protein foods--which is my catch word for meat and fish and chicken and various vegetarian products--and other snack foods, snack foods other than potato chips and French fries.

Okay. I have to cover some caveats before I show you the data. I'd ask you to keep in mind that these data cover a limited number of food categories. Obviously, the categories I show don't cover all foods. Within those categories we've only tested a limited number of products. And for those products, we may have just tested one brand or perhaps two brands.

Also, these data don't address in any significant way lot-to-lot variation or unit-to-unit variation, and these could be important sources of variation in acrylamide levels.

For those reasons, we consider these data an initial set of data, and we know that we need to test more samples, and as Drs. Canady and Troxell have explained, we intend to do so.

One more caveat. Data on acrylamide levels alone don't indicate exposure or risk. We also have to ask how many grams of food there are in a typical serving and how much of a food do people eat.

Okay. I'm going to first show three graphs explaining the data graphs, and then I'm going to show three data graphs. And this is the first of the explanation graphs.

I'm going to show the data in scatterplots, and--if I can get this to work. What you are going to see is acrylamide on the vertical axis. The food categories will be on the horizontal axis. And each point above the category names will show an individual result.

Now, in some cases, if some of the foods in the category have the same result or near the same result, they're going to overlap on the graph. So the number of products is given after the category name. For example, Category B has a 3 after it for three products. It has three dots. Category A has a 4 after it. It only has three dots showing because two of those points are the identical value.

Okay. As I mentioned, again, I'm going to show three data graphs, and I just want to point out that the categories are grouped on the graphs by the number of products tested. So the first data graph has the categories with the most individual products. The last data graph has the category with the least number of products. There's still one more explanatory graph.

I just want to point out that when I graphed these data, detections that were below the limit of quantitation of 10 parts per billion were graphed as 5; non-detects were graphed as 0.

This is the first of the scatterplots showing acrylamide values by food category, and this graph includes baby foods with 23 products; French fries with 16 products; potato chips with 16 products; infant formulas, that's actually 12 products; the protein foods with 12 products; and various breads and bakery products with 12.

On this graph, French fries includes both fries that were purchased at a restaurant, probably cooked in oil, and fries that were purchased at a supermarket frozen, and then baked according to manufacturers' directions, and then tested.

Obviously there is some variability between the categories here. The French fries and potato chips, as we've heard from other countries, have come out higher than other categories.

It's worth noting that infant formulas, you can see it really looks like there's one dot there on 0. All of the infant formulas came out below the limitation of quantitation of 10 parts per billion, and there were two that were--10 of the 12 were non-detects; two were barely above that level.

Take a look at the French fries and the chips. These are categories with some higher values, but you'll also notice that there's a lot of variability in these categories. There are a large number of lower points, and what this says to us is that there are ways of making these products, whether it's by processing techniques or ingredients, that can minimize the amount of acrylamide formed, and I think that's an important point for the future when we think about ways to reduce acrylamide in foods.

The second graph shows cereals with 10 products; other snack foods with 10 products; gravies and seasonings, 9 products; nuts and nut butters with 7; and crackers with 7.

There's not as striking a difference between these categories as in the last graph. It is worth pointing out that gravies and seasonings, that was something that our food scientists suggested we look at. Actually, it turned out to be almost completely on the low side. There are nine samples, eight of which were non-detects.

The other categories, again, we see that there is a fair amount of variation within these categories, and, again, that brings us back to the idea that there are ways that some of these products can be made that will minimize acrylamide formation, and that, again, is something we want to keep in mind for the future.

Okay. The final data graph in this series has chocolate products; canned vegetables, 5; miscellaneous is 5; cookies, 4; and ground coffee, 3. And while you look at the array of points, I'll just point out that the ground coffee is coffee before brewing, so you would likely have a different result in a cup of brewed coffee. And I don't believe we've run an analysis on that yet.

Also, the chocolate products, most of those products are actually products used as ingredients in baking. So, again, the final value when a product is consumed would be different than what you see there.

Okay. We're also doing mechanistic research, as Dr. Canady mentioned, looking at such factors as time and temperature and ingredients. These data are more preliminary, so we're not really at a point to expound on it at great length. I just wanted to show you one example of the types of factors we're looking at.

This graph shows the additional cooking, effect of additional cooking on various bread products. As you can see in the title, we're saying that additional cooking has variable effects on acrylamide levels in purchased bread products. The products on the left are products like, say, bread that you buy that hasn't had yet any additional cooking. On the right, this is what happens after additional cooking. For some products, acrylamide levels have gone up. For others, they have stayed pretty much the same. And really, the only conclusion we can draw from the slide at this point is that cooking has variable effects, and something we need to learn more about is what explains these differences.

To summarize, we have measured acrylamide levels in a broad range of products. Acrylamide was detected in a variety of foods and at a variety of levels. In some foods, no acrylamide was detected.

We are finding variability between food categories, within food categories, as well as between different brands.

We need to test more and different kinds of foods, as specified in our action plan, and I just mentioned briefly research on mechanisms. It's preliminary, but it's begun. And more mechanistic research is needed as well, as specified in our action plan.

That concludes my talk, and I'll turn it back over to Dr. Troxell.

[Applause.]

DR. TROXELL: I, too, would like to just re-emphasize the importance of the fact that the data is limited. On some samples, we literally have just one data point.

You know, in the old days when we were kids, we used to do connect the dots. Remember with connect the dots we could probably, with a few data points, either be connecting dots that turn out to be an elephant, a tiger, or a turtle, you know, or whatever. So, again, we need to be very careful that we don't jump to conclusions about what the true distributions are, and that's the importance of the additional work, and hopefully through the efforts of the multiple stakeholders and the JIFSAN clearinghouse we'll compile quite a database, and we'll be able to describe these distributions well.

Nevertheless, I think you can see from Dr. Posnick's presentation that Dr. Musser and his group have done an amazing job in developing a method and performing well over 500 highly specialized analyses between analyzing samples and determining the performance of the method. And I particularly would like to thank him and his group for the enormous effort they made on this since early June.

Again, as Dr. Posnick has indicated, there's a great deal of variation in the results from sample to sample. Variation with degree of cooking is not understood. If acrylamide can be limited by not cooking excessively, what does that mean? You know, what does that mean for toasting, for fries, for hash browns, for baking breads, for other cooking processes?

We need to understand some of these factors better and need to integrate factors relating to adequate cooking to assure foodborne pathogens are destroyed and need to integrate factors relating to nutrition, such as consumption of fiber-rich foods.

Therefore, our initial consumer message is that consumers are advised to eat a balanced diet, choosing a variety of foods that are low in fat, rich in high-fiber grains, fruits and vegetables. This is consistent with dietary advice that has been given by the government and health professionals for years and is consistent with the food pyramid. Many people would benefit by adhering to this advice.

As more information becomes available, we will consider additional messages related to cooking and foods consumers may want to limit to reduce exposure. We will want to assure that the messages will be effective and will contribute positively to public health.

With this, the morning session is concluded, and it is now 11:45. We have a wide variety of restaurants, that is, Lou's Cafe out front, nearby. The area is growing.

[Luncheon recess.]

A F T E R N O O N S E S S I O N

[1:03 p.m.]

MS. OLIVER: Good afternoon. I'm glad to see that most of the people have already made it back, and pretty close to on time, too, so thank you very much for that.

Several people have asked me if I'm going to do a half-hour presentation at the start of this afternoon, namely, our two afternoon speakers. I'm just going to introduce our afternoon speakers, and then I'll summarize the meeting at the end.

So, this afternoon, we have two speakers before we go into the public comments. Our first speaker is Dr. Michael Jacobson from the Center for Science in the Public Interest. Mike?

DR. JACOBSON: Thank you very much, Janice, and I wanted to express my gratitude to the Food and Drug Administration for allowing us--first, for holding this meeting, and then allowing me to be a speaker this afternoon.

I think the attention that the FDA is paying to acrylamide is well merited. Acrylamide is widely recognized to be a carcinogen in rats and mice and, based on those studies, is considered to be a probable human carcinogen. The epidemiological data on acrylamide is very limited, though the major study linked acrylamide exposure to pancreatic cancer in workers.

Some industry spokespersons have been saying that the amounts of dietary acrylamide are trivial. However, the government of Sweden, a country with roughly one-thirtieth the population of the U.S., estimates that the contaminant may cause several hundred cancers a year over in Sweden. That suggests, assuming similar exposure in the U.S. and Sweden--and, frankly, I think that we could beat them any day in potato chip and French fry consumption---that acrylamide may be causing at least several thousand cancers per year in the U.S. Multiplying that figure by the average American's lifespan, about 75 years, suggests that acrylamide may be causing cancer in roughly 200,000 people out of a current population of 280 million. While that is not in the league of tobacco, nowhere in that league, it is hardly a trivial number.

Acrylamide is also well recognized as a neurotoxin, and the amounts of acrylamide that we are consuming in foods may be significant. Last June, the Food and Drug administration "determined that the acceptable daily intake of acrylamide with respect to neurotoxicity to be 12 micrograms per person per day..." Twelve micrograms.

Considering the paucity of research that could identify--neurotoxicity research that could identify a true no-effect level--the ADI of 12 micrograms per day may be overly generous. Nevertheless, the average American appears to be consuming several times that much in French fries, potato chips, and other foods. Heavy consumers of such foods may be getting 10 times as much acrylamide as the FDA's acceptable daily intake.

To begin giving Americans some sense of how much acrylamide is in some common U.S. foods and to contribute to the general pool of knowledge about acrylamide, earlier this year CSPI commissioned the Swedish Government to test about a dozen foods, and then we provided the test results to the WHO conference in June that you heard about earlier, as well as to the FDA and general public.

In this slide, if you can show the first slide--am I supposed to do anything here? No? It's got to warm up. I didn't put you to sleep already, did I? Good. Thank you.

This slide shows our results. We've expressed the acrylamide levels in micrograms per serving rather than per kilogram to give consumers a better sense of what they would actually ingest when they eat a serving of food.

We think that brand-name information is very valuable in two ways: First, consumers can base some choices, if they see two similar foods, one has 10 times as much acrylamide as another, they might change their buying decision. Also, brand-name information is a powerful incentive to companies down there at the bottom end of the scale to change their practices.

So the numbers are very similar to what the Swedes found, obviously, French fries being the greatest contributor, snack chips being somewhat less on a per serving basis, and a number of foods having relatively little.

The next slide, please?

It's also worth comparing acrylamide levels to the amount of acrylamide that might be permitted in California without a Proposition 65 warning notice or restriction. Individual categories of food exceed California's limit, as does the sum, the daily sum of those foods. And these estimates are based on--the French fries were based on the CSPI test data and the rest of that information is based on the Swedish data using California's limit on acrylamide, which it set some years ago. So it's quite clear that if you use California's limit, which is based on one cancer in 100,000 people, the amount in foods, these different categories of foods, really, those levels are really quite significant, not something that can be brushed off as, ah, just a few micrograms.

That's enough for the slide, please.

I fully support the FDA's Action Plan, particularly its efforts to conduct its own research and monitor and stimulate research by industry and academic scientists. It's important that this research explore the amount of acrylamide in various foods, the chemical's toxicity, the chemical reactions that lead to acrylamide, and ultimately figure out ways to prevent the formation of acrylamide.

The action plan does not mention neurotoxicity in particular, but I hope that the government will conduct long-term animal studies to identify potential subtle effects at lower levels than have been studied so far. It is exciting that early research by Health Canada is already paying dividends. The Canadians recently found that acrylamide can form from a reaction between the amino acid asparagine and glucose, and that obviously points to ways to try to change, reduce acrylamide contamination by changing the potatoes, perhaps, the variety of potato or the storage.

I'd like to see the FDA give consumers useful information and advice related to acrylamide. Its past advice is typified by this statement, which came from the Federal Register notice three weeks ago: "Until more is known, FDA is not recommending that consumers change their diet or cooking method because of concerns about acrylamide." That isn't helpful, and it isn't consistent with the FDA's stated goal of informing and educating consumers. The FDA should tell the public that acrylamide poses a real risk and should provide information about acrylamide levels in the individual foods that it tests. Where there are significant differences between brands, that information could lead to smarter consumer choices and also to improvements by companies marketing the more contaminated products.

Having the general product categories is--I don't mean to belittle that. That's very important information for the public. But we'd like to see the FDA be more forthcoming about the actual products that are being tested.

I think the FDA should advise consumers to eat less of the most contaminated, least nutritious foods--namely, French fries and snack chips. Obviously, most people would have done well to eat less of those foods long before acrylamide was discovered in them, but now people have yet another reason to eat less of them, and the FDA, the leading public health agency in the United States, should be giving that information straight out to the public.

Thank you very much.

[Applause.]

MS. OLIVER: Thank you.

Our next speaker is Dr. David Lineback, Director of the Joint Institute for Food Safety and Applied Nutrition here at the University of Maryland and at FDA.

DR. LINEBACK: Thank you and good afternoon. As Director of the Joint Institute for Food Safety and Applied Nutrition, or JIFSAN, as it's called, at the University of Maryland and Chair of the Scientific Advisory Panel for the American Association of Cereal Chemists, I have had the opportunity to not only stay informed but also participate in many of the national and international scientific discussions that are ongoing with respect to the occurrence of acrylamide in foods.

The FDA draft action plan indicates that JIFSAN will serve as the clearinghouse for continued international coordination and data sharing. We have been asked by the FAO/WHO to manage the international Acrylamide in Food Network that is being established as a result of one of the recommendations from the FAO/WHO Expert Consultation in June. The Web-based network will invite all interested parties to share relevant data as well as ongoing investigations. We are actively engaged in this effort and strongly endorse the need for ongoing coordination and information exchange nationally and internationally. I think it is important to keep in mind that this is clearly a topic that has been and will continue to be an international issue with efforts underway from governments, academia, and industry.

FDA is to be complimented for developing a thoughtful and very appropriate action plan for addressing the occurrence of acrylamide in food. I believe the agency has identified a course of action that takes into account what is known or, more precisely, what is not known about acrylamide in foods and is consistent with the activities and directions being taken by national and international scientific bodies and governmental organizations.

Given that the issue of acrylamide in food only surfaced in late April of this year with the press announcement from Sweden, the progress and level of activity achieved by scientists and government organizations in the United States and other countries is truly outstanding. With the excellent analytical work of the Swedish scientists and the recognition of the potential serious implications, the decision was made to release the findings via a press conference only a short time after the revised paper describing the investigation had been accepted for publication. This attitude of sharing is now characterizing the approach being used by scientists in many nations.

For example, the Central Science Laboratory, who did the analyses in the U.K. and with whom JIFSAN has a cooperative relationship, shared details of the analytical methodology and actually some of the samples that they had analyzed with us. This was then shared with FDA and other analytical groups within the U.S. Preliminary data are released and shared before the months required for publication in a scientific journal. This is a different approach for science, one that is often criticized, but this is moving the scientific investigations forward rapidly and emphasizing collaboration in approaching this complex issue.

With the rapidity with which this new and important information is being acquired and shared, I believe that care must be taken not to reach scientific, policy, and regulatory conclusions based on the most recent press statement, media release, or shared information. Plans for rapid, systematic approaches to obtaining essential information, such as FDA has developed, are being prepared in other countries, and there is clear evidence that needed work is being initiated as quickly as possible. Decisions about what, if anything, must be done relative to acrylamide in foods should be made as quickly, but also responsibly, as can be done using sound, valid scientific information.

I want to share several observations, particularly with regard to what is occurring internationally. At the outset, I think it is the consensus among scientists, governments, and industry that the new and unexpected information provided in the announced finding of acrylamide in foods from Sweden and subsequently in other countries calls for the concerted effort that is underway to gather additional and new information needed to characterize the potential risks posed by acrylamide in food and determine what are appropriate and effective responses.

These informational needs can be grouped into four major areas: first, the determination of the amount and extent of acrylamide in foods; second, the determination of the mechanisms involved in its formation in foods; third, human exposure and biomarkers; fourth, toxicology and metabolic consequences.

This approach is being reflected in the meetings, discussions, and initiation of research internationally.

In terms of data on acrylamide in foods, the data collected in Sweden and subsequently in a number of other countries, including the United States, the U.K., Norway, and Switzerland, confirm the presence of acrylamide in selected types of foods, as we have seen again illustrated. This general finding has been sufficient to establish acrylamide in foods as a concern and to galvanize action at the international and national level.

However, the limited data available on acrylamide in foods has also demonstrated that much more information is needed as a fundamental first step. National and international bodies recognize that the full range of affected food types has not been determined. The extent to which acrylamide is found in home-prepared foods, in particular, remains a significant unknown. Similarly, within specific food types, considerable variation has been found in the levels of acrylamide.

I am aware of no country or international body that has used the limited information available on levels of acrylamide in specific food types to take targeted regulatory action and, to my knowledge, only one has issued a consumer advisory on specific product types: Norway's National Food Authority recommended "that people who eat a lot of potato chips in particular should consider to cut back for additional safety."

The limited data available and the consequent uncertainties about the levels of acrylamide within and across different foods contribute to the position taken in the June FAO/WHO consultation on acrylamide in food and the July opinion of the European Commission Scientific Committee on Food that existing general advice to the public on healthy eating remains valid and appropriate. Individual countries have also taken this approach when considering what, if any, advice is currently appropriate. The language used by the EU Scientific Committee on Food is: "The information available on acrylamide so far reinforces general advice on healthy eating. People should eat a balanced and varied diet, which includes plenty of fruit and vegetables, and should moderate their consumption of fried and fatty foods." I am not aware of any country, other than the ones I cited, or international body that believes enough information is available to give advice to consumers on individual products or categories of product.

In terms of the mechanisms of acrylamide formation, a great deal of attention is being given nationally and internationally to determine the mechanisms for acrylamide formation in foods. Temperature and time are two factors that are consistently identified. Given these important factors, however, various ways of preparing food, with the exception of boiling, have not been eliminated as contributing to acrylamide formation. In fact, recent evidence tends to indicate that it is the actual temperature and time, rather than the method, which contributes to acrylamide formation in food. Knowing the mechanisms of acrylamide formation, including possible reactants and precursors, is critical information for identifying possible ways of reducing or minimizing formation. There is recognition that home and commercial food preparation methods need to be considered. For example, during the information seminar held by the German Federal Institute for Health Protection of Consumers and Veterinary Medicine, it was noted, "A special problem is posed by the preparation of deep-fried, baked, and fried foods in the home, mass catering facilities, and restaurants. Experiments show that this critical substance is also formed there."

Ongoing research dealing with the mechanisms of acrylamide formation reinforces the need for FDA and others engaged in dealing with this issue to stay current with rapidly developing information and not to commit to a course of action that may prove inappropriate.

On September 11, Health Canada issued a letter that describes model experiments that found when asparagine--the most predominant free amino acid found in potatoes--is heated with glucose, which is a "reducing" sugar--also present in potatoes--acrylamide is produced. French fries made with some potatoes had only one-fifth the acrylamide levels of others, because the potatoes were initially lower in asparagine. Last week, at an Association of Analytical Chemists meeting, work done by Procter & Gamble was reported showing that the amino acid asparagine, coupled with a carbonyl source like glucose, is a major precursor of acrylamide in food products. In fact, the carbons and nitrogen in acrylamide were found to come from asparagine, with the source of nitrogen in the acrylamide being the asparagine amide nitrogen. These studies, yet to be published, may be significant in finding ways to affect acrylamide formation and in identifying what foods may be of interest. However, being able to reduce the amount of acrylamide formed does not indicate what the effect will be upon food safety or the actual nature of the food product. These reports demonstrate that mechanistic research is underway and that new information will likely be surfacing at a rapid pace.

In determining the possible risk to the public, while national and international organizations are taking the presence of acrylamide in foods very seriously given what is known about the compound, there remains major scientific uncertainties and the need for research to determine if and what the health risk is from acrylamide in foods. Since acrylamide is apparently formed during food preparation at relatively high temperatures, we must remember that we have most probably been exposed to acrylamide in foods since we started using heat in food preparation. Statements from national and international organizations repeatedly emphasize that there is insufficient information to determine what the risk is. Similarly, there is not agreement on whether or if a quantitative risk determination for acrylamide in food is possible.

For example, the need for research to address the question of whether acrylamide, and specifically acrylamide in food, causes human cancer has been described. The FAO/WHO consultation acknowledged that quantitative human cancer risks using experimental animal carcinogenicity data have been attempted using different models. However, the consultation did not reach consensus on how quantitative risk assessment based on animal data should be used to estimate human cancer risk from acrylamide in food. Clearly, determining and, if possible, quantifying human risk is of interest internationally, but without additional research and resolution of scientific uncertainties, there does not appear to be a consensus on this issue within the scientific community or among government organizations.

In terms of the relevance of FDA's plan of action to international efforts, I'm going to indicate that acrylamide in foods is definitely an issue receiving national and international attention. In my opinion, FDA has developed an appropriate and aggressive plan of action that is consistent and in keeping with other national and international efforts. The EC Science Committee on Food recommended initiatives for research related to the following:

First, the mechanisms of formation of acrylamide in food; the levels in food and extended dietary exposure assessments, covering also national variations; bioavailability of acrylamide in food; elucidation of the mode of action as a carcinogen; investigation of the relationship between dietary intake of acrylamide and formation of glycidamide-DNA adducts; analysis of dietary acrylamide intake, exposure biomarkers, and disease endpoints in existing European and worldwide epidemiological cohorts; epidemiological studies on cancer in populations of known high exposure, such as occupationally exposed workers.

These priorities recognize that much more scientific information is needed to meet the recommendation of the committee that levels of acrylamide in food should be as low as reasonably achievable. I believe the FDA proposed plan of action is definitely consistent with the steps being taken by the EC Science Committee on Food and other national and international bodies and will make a significant contribution to this global effort.

Thank you for the opportunity to share these remarks with you. I will attempt to answer any questions you may have.

[Applause.]

MS. OLIVER: Thank you, Dr. Lineback.

That concludes the formal presentations for today, and so now what we're going to do is take public comment. And you'll see that there are two microphones at about the fifth row of seats, one on the left side and one on the right side of the auditorium. So I'd ask those who are going to make public comments if you would come down. I have two people that are signed up so far. If you would rather make it from where you're sitting, we can have somebody bring a microphone up to you.

The first presenter, first oral comments, are from Sue Ferenc from Grocery Manufacturers Association. No? You're not going to make comments, okay.

The next is from Dr. Maureen Storey, Acting Director and Research Associate Professor from the Center for Food and Nutrition Policy. Would you like to come down to the microphone, or would you like to make your comments from your seat? Okay.

DR. STOREY: [Inaudible - off microphone] -- foods. Finally, Dr. Sanford Miller from the center offered some insights on lessons learned from past experiences, such as nitrosamines found in some alcoholic beverages.

Why did it take so long to discover acrylamide in food? It's because until now no one was looking for this particular compound. Carcinogens and other naturally occurring toxicants like mycotoxins are present in our foods without a doubt. Nevertheless, acrylamide is not known to occur naturally in the environment, but it is not unique with regard to other compounds formed during cooking.

Cooking foods to enhance flavor and provide protection from foodborne pathogens can lead to the formation of acrylamide and other carcinogens such as polycyclic aromatic hydrocarbons and heterocyclic amines. And there are far greater environmental risks from other non-food sources of acrylamide exposure. For example, smokers have nearly four times the internal exposure to acrylamide than non-smokers as measured by hemoglobin adduct formation.

To date, animal studies are unclear as to cancer risk from acrylamide. In cancer studies using very high doses of acrylamide in drinking water, male rats receiving the highest dose of acrylamide have significantly more benign tumors of the thyroid and testes, but no more malignant tumors of the thyroid or central nervous system and spinal cord than control rats. Female rats receiving the two highest doses of acrylamide in drinking water developed more benign tumors of the mammary gland and thyroid, but no greater incidence of malignant thyroid tumors than control female rats.

As has been stated earlier, there is no epidemiological evidence that acrylamide is a human carcinogen. Cohort studies conducted in the 1980s showed no increase in cancer rates among industrial workers exposed to acrylamide through the skin and by inhalation. Once acrylamide enters the body, we know the biochemical pathways that metabolize and excrete it. But with regard to exposure through foods, we do not know how much acrylamide is actually excreted without absorption by the gut.

Perhaps the single most important thing we learned during the roundtable was that we know very little about the human exposure to acrylamide in foods and whether there is a cancer risk to humans or not. The data on acrylamide concentration in foods are few.

What the data do not tell us also is how much acrylamide is absorbed by the gut, how much of the absorbed dose is metabolized and excreted, and how much forms hemoglobin adducts or DNA adducts, and then how much is required to causedamage that eventually may lead to cancer.

This exposure assessment is also limited by the weak data on acrylamide concentration in foods. As new and better data on acrylamide concentration in a variety of foods become available, estimates of exposure through food could change substantially. This is particularly true since the current data on acrylamide is highly skewed and heavily influenced by only a few products with highly variable concentration levels.

The center applauds FDA's Action Plan to conduct research that will shed more light on a number of aspects related to acrylamide in the food supply. FDA should take this opportunity to develop, as they have done, a comprehensive plan that includes all elements of a risk analysis, risk assessment, risk management, and risk communication.

Future research will undoubtedly identify other compounds in foods that may or may not pose a health risk or pose difficult regulatory problems. It is, therefore, critical that the agency implement an action plan efficiently and effectively to protect the public health, if the risk is there, and yet maintain an adequate food supply.

Thank you.

MS. OLIVER: Thank you very much.

That's the number of speakers that we have signed up, but we're really interested in public comments, and there are a lot of you here. We're interested in your thoughts on the action plan, so do any of you have any comments you'd like to make? And if you would state your name and the organization you're with?

DR. CHIN: I'm Henry Chin with National Food Processors Association. NFPA--National Food Processors Association--appreciates this opportunity to comment on the FDA Draft Action Plan for Acrylamide in Food.

As has been mentioned before, NFPA and our member companies are closely following the developments on the issue of acrylamide in foods and wish to take this opportunity to express our desire and intent to continue working with FDA in pursuing relevant scientific investigations and taking appropriate actions based upon what is learned during this period of active research and information gathering.

NFPA agrees with FDA that questions and possible concerns raised by the new findings of acrylamide's presence in foods by the Swedish researchers need to be addressed and that a deliberative and scientifically sound approach is essential in assessing and appropriately addressing any public health issues.

We believe that the priority areas identified for FDA for immediate action are appropriate, and these include the developing of rapid validated methods for analyzing acrylamide in foods, the collecting of data on levels of acrylamide in food, determining mechanisms of formation, and, of course, the critically needed information for informed risk assessment.

While NFPA agrees that many questions regarding finding acrylamide in foods must be addressed expeditiously, we do not believe that we are faced with a public health emergency. The recent results reported by the Canadian Government and also by the researchers at Procter & Gamble about a potentially important mechanism for formation of acrylamide strongly support the notion that there is a long, long history to the human consumption of acrylamide in a variety of foods. Thus, the potential national and international implications call for a deliberative and scientifically sound approach to addressing this issue, and we applaud FDA for laying out such an approach.

FDA's draft plan suggests science should lead the way in determining what actions are appropriate, and we wholeheartedly agree with what FDA has laid out.

MS. OLIVER: Thank you.

Does anyone else have any comments?

[No response.]

MS. OLIVER: Are you sure you have no comments?

[Laughter.]

DR. BERNARD: Bruce Bernard of SRA, a consulting firm in Washington. We would like to go on record as applauding the FDA for the approach that it's taking. Although Americans--humans in general and Americans in particular are quite enthralled with problems and things that scare them, and certainly newspapers love to make use of that, we think that the deliberate methodological approach that FDA is using is very important not only to secure the truth of what actually is out there in terms of the science, but also as a template for what surely is coming. And what is coming is, of course, as we are able to detect compounds in food and other places at lower and lower levels--I think now we are halfway to Avogadro's number--we are going to find more and more things there that we never expected.

And what we really need to do is to set up a methodology, an approach, a rational approach to evaluating public health when new information comes available in the absence of a brouhaha that I think would be supported by both politicians and the media.

So I hope you stay to the plan and stay the course. Thank you.

MS. OLIVER: Thank you.

Any other comments?

[No response.]

MS. OLIVER: Okay. Seeing none, let me try to summarize a little bit from this morning and from this afternoon.

I think when we started off, you know, we started off to present our action plan, and we presented some results, and the goal of our overall plan is to prevent and/or reduce the potential risk of acrylamide in foods to the greatest extent feasible. And I think you heard that over a number of times.

We have talked about, too, and a number of our speakers talked about this afternoon what is new and what is not new. What is new is that we're now able to detect acrylamide in the foods and that there's knowledge that cooking can affect the levels of acrylamide. But there's also some that is not known about that. So acrylamide is not something that has just suddenly appeared in the food. It's just that we're now finding it out.

Our draft action plan included a number of sections, and I'll just go over them slowly.

One was to develop laboratory methods to measure the acrylamide, and that we've done and are in the process of validating.

The second is to survey acrylamide levels in foods. We've done an exploratory survey to basically frame out what foods we need to do next.

We need to examine how acrylamide is formed in foods, and as you've heard from a couple of our speakers this afternoon, there has been some additional research, even very recently, talking about a possible mechanism.

We need to identify a means to reduce acrylamide levels, and I have heard that a number of times, and no disagreement there.

We need to evaluate any human health risk of consuming acrylamide in foods, and we've heard some discussion of that this afternoon from our commenters.

We need to conduct outreach and education for both producers and consumers as needed, and in the comments we heard this afternoon, there were some differing opinions on what information should be provided consumers, and we need to take those opinions and look at that for the future and see what we need to do. All opinions are valuable.

The last point was informing stakeholders on the plan progress and any steps we need to reduce risk. This is one of our first steps in putting out the plan, getting the information to you, and trying to get comment back. And as we said earlier, we'll be accepting comments until the end of October, October 30th, and we do hope that you will provide comments to us on whether you agree with the plan, whether you think there are additional things we should be doing. We're interested in this feedback.

And as was said earlier, this is the first of a series of public meetings that will be had to gain your input into the action plan.

You've heard comments this afternoon and comments this morning on what is the risk. We know that it is an animal carcinogen, but the actual human health risk of consuming acrylamide is not known. Along that line, you've heard from CDC on some of the nature of what they will be doing to try to help us in this area.

And as has been said by a number of folks this morning and this afternoon, this is not a problem we can deal with ourselves. This is a problem that we need to deal with and we need to work with our federal partners as we've gotten together already and started research and finding out what they were doing and their assistance, to work with our international partners where we've already started in the WHO/FAO consultation, and plan to continue to do more in the future, to work with academia in many ways as through JIFSAN and through our National Center for Food Safety and Toxicology, but also a number of other academicians in the future, to work with our consumers and our consumer partners, and Dr. Jacobson represents some of this group, and also to work with the food-processing industry. I think it's something that we all have a part to lend to this.

This morning, we heard that we've done some analyses, and we are measuring acrylamide levels in a wide variety of foods. And you saw some of that variety of foods. But based on our exploratory survey, we're finding acrylamide levels in a number of foods, but we're finding a lot of variability in our results. We're finding wide variability both in food groups and across the food groups, between food groups, and within individual food groups. So we really need to do more testing to understand the actual levels of acrylamide in foods.

And as we said this morning, our advice to consumers has been to eat a balanced diet, to choose a variety of foods that are low in fat, rich in high-fiber grains, fruits and vegetables. And I hear, Dr. Jacobson, your concerns, and we will look at that. I hear other concerns also.

But over the next six months, we expect to gather additional information. We know that industry and others are already gathering additional information, and more things are moving forward. But we will be having a meeting of our Advisory Subcommittee on Contaminants in early December, and we'll be having a full meeting of the Advisory Committee in March, which will take these issues and additional data that comes forward.

We'll also continue to inform our stakeholders, either through public meetings or through press. And we really appreciate your coming to this meeting, your input you've already provided, and your input that I'm sure you're going to provide to us by October 30th.

The other thing I'd like to say is that we have presentations available for those who did not pick them up. The presentations that were made this morning are available on the tables outside the auditorium. They will also be placed in dockets and on the Web, probably by the end of this week. We will also be posting data that we've obtained on the Web around that same time.

The last thing I'd like to do is to thank everyone who has attended the meeting, everyone who has spoken at the meeting--we really appreciate your comments--and all of those who helped develop this meeting and get it ready and make the presentations. It really took a lot of work, and I really do appreciate it.

There are also, for all of you, refreshments probably in the back, so as you're leaving, you may have refreshments, stand and talk to one another, discuss this, and maybe you can think of some additional input you can give us that we'd like.

Thank you very much.

[Whereupon, at 1:44 p.m., the meeting was adjourned.]