• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Assessment of Risk: Drug Residues

Kevin Greenlees, Ph.D.


DR. GREENLEES: I want to start by thanking Dr. Rulis for laying some very nice groundwork for this talk which was -- we didn't coordinate this. We really didn't work ahead of time. And it is just -- he set such a very nice basis just to make life much easier for me.. : This talk is really to talk about how we evaluate the risk for the chemical residues ---. It is not going to address anything towards the purpose of this meeting which is a risk assessment or the safety of the consumption of a resistant microorganism.


When we are trying to put this in a framework of a risk assessment type approach, the evaluation of new animal drugs just like food additives evolved before the current concepts of risk assessment and how you do risk assessments. But I think you will find that the approach really still fits a lot of the paradigm once you have looked at all of the boxes you need to fit and all the pieces that need to fit together.

It is going to deal with problem identification, the identification and characterization of the hazard that you have concern for, how large is that hazard, what is the acceptable risk level that you are trying to deal with, what is the exposure and how are you going to deal wit that exposure. And I am going to admit right up front that I am going to mix in this both management of the risk and assessment of the risk because I think that gives you a more complete picture of how we ---.


The problem that we are dealing with is exposure of the human consumer to an unsafe chemical residue of the new animal drug. We heard definitions given of what residues are. And I am going to just briefly tell you that the residue that we are dealing with is the residue of new animal drugs, any added substance that is present in or on the commodity or food primarily as a result of metabolism or the degradation of the new animal drug.

So in other words, it is the drug you administer and all its by-products. You are going to get more about the definition in the next talk by Dr. Tollefson. I am going to leave it there, again, emphasizing we are talking about the chemical residues.


When we are talking about what is the hazard, what is your concern, we need to identify what we actually have concern for. Is it the active ingredient Is it the metabolite in the active ingredient Is it a second or third order metabolite Where does your concern lie

And there are a whole battery of studies which were alluded to in the previous study -- in the previous talk on what are those -- how do you go about doing that. So there are oral toxicity studies which are the standard, you know, rodent assays, 90-day studies, the development toxicity study, reproduction toxicity studies and any special studies that are needed to address the specific nature of that compound.

And for a given compound, you may look at that and decide we don't need one of those studies, again, because of the nature of that compound. The -- in addition to identifying what is it in the residue that we have concern for, what are its characteristics, we are also assessing what is the nature of that toxicity. It is developmental toxicity Does it cause birth defects Does it cause damage to the liver What kinds of toxicity do we see

We look specifically at the carcinogenic potential of compounds. This is done through assessment of the oral bioassays, the literature and through looking at specific immunogenicity assays. And if necessary, we will go on to make quantifiable assays where we actually are looking for development of cancer in animals.

In recent years, we have started to look at the effects of the drugs residues, the chemical residues in food on the human gut and flora recognizing that it is possible that it may have effects on the gut and flora at lower levels or different levels than you might see in systemic toxicology where you have effects, again, on one of the organ systems of the body indirectly.

And also part of the same characterization process is the residue studies that are required. And I listed here the --- metabolism study because that is probably the big gun is usually where you typically will do a carbon-14 study looking at what are all of the residues, where do they go and what are all the pieces that are in the animal.

But it also is based on other studies which will again characterize the nature of the residues that are in the animal. And between those studies, then you can get a handle on where you have your level of concern and what concern that would be.


When you have completed that process, you then have to go through each of those studies, evaluate what that information is. And as was again talked about previously, our typical approach is to try and establish a dose to the animal in the animal studies that would have a no observable effect level.

In some cases, you may actually have an effect. And that would be a --- effect level. There are also other approaches such as the benchmark dose which allows you instead of just saying, well, what dose do you not see an effect, it allows you to use the dose response relationship from those doses we see an effect and calculate back to a -- the level which is comparable to a low effect level.

These in turn allow us to calculate or determine acceptable daily intake. And there are other end points such as the reference dose or safe concentration which are also -- again, these just come to say how much are we going to allow in the diet. That's what these numbers mean.


There is some difference to how you deal with a carcinogen than a non-carcinogen. For a non-carcinogen, we -- as I mentioned, we go through the sub-chronic toxicity study, reproduction, development toxicity. It goes through safety of gut flora. You evaluate all of those studies, establish the NOEL and you eventually come up with an acceptable daily intake.

If a compound is thought to be potentially carcinogenic, then we also have to evaluate its potential to cause cancer in rodent bioassays. We also make look at some alternative assays. But these are assays the studies insist we design to show does it cause cancer and can we establish a dose response relationship to that cancer.

If the answer is, yes, it does cause cancer and you can then still calculate a dose relationship, then our approach is to determine a one in a million risk level and test --- using a linear low dose extrapolation. We calculate what is the equivalent of acceptable daily intake and the equivalent of safe concentration. So it just builds on what you do for a non-carcinogen.


The way to use that is you have to have some standard. The standard is the same as that for food additives, that it is the reasonable certainty of no harm. For carcinogens, this is the upper bound of the dose resulting in a one in a million risk level -- the dose -- this can be interpreted as saying the safety of no harm is that dose which will have a risk of one in a million of causing cancer in the rodent that you studied it in.

And the reasonable certainty is established by having the upper bound, the variability accounted for in that calculation.

For non-carcinogens, it is based on the acceptable daily intake which uses safety factors as we talked about before and to calculate a no effect level.

Unlike food additives where they have an estimated exposure daily intake or an estimated exposure, we assume that all the animals are medicated with the drug. So if this is a drug intended for dairy cattle, we assume all dairy cattle are medicated. We then assume that all of the edible tissues are at the maximum allowable concentration, the tolerance concentration.

We then also -- and this is not on the slide. We also then assume that people are going to consume a specific quantity of that daily. So for muscle, it is 300 grams. For liver, it is 100 grams and so on. The latter -- the consumption factors are based on data, based on --- surveys. The other assumptions are just that, they are assumptions. But they are very conservative assumptions.


When you have all that information, we then -- because we are working to a standard to say we are going to have to meet that end goal, we then use other studies to help us get there. So we look at the drug metabolism studies to say, okay, we are going to want to measure this compound. So you have to -- and we did all of our toxicology based on all of the residues.

We can't measure all of the residues. You've got to have something you can actually get a handle on. So we establish a marker residue which is something that you can actually measure by an assay and say what is the relationship of that marker residue to the total residue. And then you actually develop the assay to go with that so that you can go and measure how much of that compound is actually in the edible tissues. And note how that refers to all the edible tissues.

We establish a regulatory tolerance which is the safe concentration which was calculated in the acceptable daily intake as is all the residues. The tolerance is what you can actually measure of that safe concentration. And you establish a ratio to that. So from the tolerance, you can take a direct line back to say what would be the safe concentration. And then you know whether you are or are not within the acceptable daily intake ---.

We calculate a withdrawal period based on withdrawal studies. So we have already established how much is allowed in the diet. If we say, okay, how long do you have to take the animal off the drug and allow the drug to deplete until you have actually reached that level. And, again, there are conservatisms in here so that we are confident that not only do you have a safe concentration in the diet and you calculate the safe amount, but you actually are confident that the animal population will have reached that dose, that concentration in the animal tissues.

And there are also other mitigating or management factors. And this is all dealing with risk management. But you notice it is also data analysis at the same time. Where you might have restrictions on the label, there certainly are indications on the label and instructions on the label all of which will assure the commodity is the drug product that is used appropriately.

And there is a post-market surveillance and compliance, again, to be sure that the compound is used appropriately and that they are following the label.

What I did not talk about earlier in the talk, but it takes place throughout the entire process, is the communication process between the Food and Drug Administration and the drug sponsor so that there is interaction throughout this entire process to ensure that we have got the best information and are really on board with what --- the compound and what would be a safe criteria for approval.

Once a product is approved, then it turns to external communication with the rest of the public. And we are dealing with -- we have the label information. You have freedom of information summaries which summarize all the bases for the approval. And then we also have communication from the drug sponsors on safe and appropriate use of the compound. Thank you very much. I am going to stop there.

DR. STERNER: Questions for Kevin

MR. : Yes, I've got one. I wonder if you would clarify or elaborate a little bit on the use of the low observed effect level in calculating the acceptable daily intake.

DR. GREENLEES: For some compounds and in some circumstances, you may, in fact, have studies which do not have a no-effect level, but in fact show a low effect level. In other words, you have actually -- the lowest dose administered has some effect.

If you look at the sum total of the data you have, you may elect -- you may determine that you can appropriate establish the safety of the compound by simply using a larger safety factor, a larger uncertainty factor. No effect levels are to some extent a -- it's a product of study design.

How close did you estimate what would, in fact, be a no-effect level dose when you designed the study You may be right there. You might have missed it. You could -- the no-effect level could be 100-fold lower than the threshold which would actually not show an effect.

So you have to evaluate the study on the basis of your sum total of information on that. But in the cases where low effect levels of used, then an increase is added to the safety factor, usually using ten-fold instead of using 1,000-fold that you would otherwise have been 100-fold safety factor.

DR. STERNER: Thank you for staying on time as a small subtle reminder to our next speaker. Our next speaker is talking about assessment of risk with regard to pesticides is Dr. Roy Sjoblad. He has been at the Office of Pesticide Programs -- see, I am off already -- Pesticide Programs at the U.S. EPA since 1984.

He is currently a Senior Microbiologist in the Biopesticides and Pollution Prevention Division. He is involved in a number of policy issues related to safety and use of genetically engineered microbial and plant pesticides.

And he received his Ph.D. from Pennsylvania State University. And prior to joining the EPA, he was a faculty member in the Department of Microbiology at the University of Maryland. Dr. Sjoblad