Zero Tolerance Concept as It Relates to Hormonal Residues in Animal Products Albert C. Kolbye, Jr. J Anim Sci 1975. 40:1258-1261. The online version of this article, along with updated information and services, is located on the World Wide Web at: http://jas.fass.org www.asas.org Downloaded from jas.fass.org by on January 3, 2010. Z E R O T O L E R A N C E C O N C E P T AS IT R E L A T E S T O H O R M O N A L R E S I D U E S IN A N I M A L P R O D U C T S 1 Albert C. Kolbye, Jr. 2 When we talk about "zero tolerance" we are essentially speaking about the requirements of the Delaney Clause which appears in the FD&C Act with respect to food, color, and animal feed additives. The concept of zero refers to the requirement that no residues of the compound be detectable by an analytical method specified by the Secretary in food products intended for human consumption if such compound has been shown to be carcinogenetic. Detectable obviously refers to analytical methodology and the question then becomes one of determining how small "zero" is at any given point in time. ltow small because obviously zero must be defined in terms of the sensitivity of specific testing methods to detect any residue present within the limits of detection. When the Delaney Clause was enacted by Congress, analytical sensitivity by chemical methods ranged in the parts per thousand and parts per million. In the ensuing period of time, analytical methods have been improved with respect to sensitivity such that a sensitivity of parts per billion and in some instances parts per trillion is within our capability. The question then becomes, how can we reasonably predict with some degree of standardization, what would be the desired level of sensitivity with respect to these compounds? Otherwise, predictability of legal use of certain compounds suffers from the vagaries involved in the advancement of scientific knowledge as new methods with increased sensitivity are developed and thus we get caught in the never ending game of chasing Presented as part of a Symposium, The Use of zeros. Since the method of testing must be Antimicrobials and Hormones for Livestock Produc- approved by the Secretary, which methods tion, at the 66th Annual Meeting of the American should be deemed approvable? Society of Animal Science at the University of Previously, FDA has been criticized by some Maryland, College Park, Jtdy 30, 1974. as having what has been characterized as a 2Associate Director for Sciences, Bureau of Foods Food and Drug Administration, Washington, D. C. reflex reaction to the sensitivity question: two parts per billion. This was done for two reasons, 20204. 1258 At the outset I should like to pause for a moment and pay tribute to Dr. Leo Friedman, Director of the Division of Toxie~)logy, who would have been speaking to you now on this topic. His recent death understandably has created a great sense of loss in the Bureau of Fooas. To talk about zero tolerance as it relates to hormones is not exactly a project of my own choice for such a talk must necessarily involve some discussion of complex mathematical models in conjunction with considerations pertaining to the etiologies and mechanisms involved with induction of cancer. Not only is our current state of scientific knowledge about the mechanisms by which cancers are induced best described as abysmally inadequate in most respects, but with apologies to my friends involved with hormones, much remains to be learned in that field too. And the interrelationships between the two fields-carcinogenesis and endocrinology - deserve much additional research attention. Therefore, my talk this morning must necessarily be more general than specific, more philosophical than substantial, and raise more questions that appear to me to be relevant than I can answer. I would like to share with you some of the background considerations as I see them and attempt to describe to you where matters stand in our thinking in the Bureau of Foods. J O U R N A L O F A N I M A L S C I E N C E , vol. 4 0 , n o . 6, 1 9 7 5 Downloaded from jas.fass.org by on January 3, 2010. ANTIMICROBIALS AND HORMONES FOR LIVESTOCK 1259 firstly, the judgment of experts in food safety which cancers are produced is still in a state that such a level was negligible in the sense of that might best be described as abysmally any substantial biological importance with re- inadequate. The Delaney Clause reads in terms spect to most compounds known and the belief of inducing cancer which then has been interthat if the risk of incurring cancer increases preted by some to mean any substance that with increased exposures then the opposite causes a statistically significant increase in must be true in that the risk of incurring cancer malignant tumors in test animals as compared obviously decreases with decremental ex- to controls is likely to be deemed a carcinogen. posures. The other reason was that in general But a relevant question immediately comes two parts per billion was about the lowest to mind: what is the biological significance as practical enforcement level consistent with cur- compared to the statistical significance? When rent capabilities of analytical chemistry, with we consider hormonal agents in relation to respect to many added substances such as DES. carcinogenesis, we are entering into complex With the advent of newer techniques prim- territory somewhat darkened by the lack of arily in research and not yet generally capable research light. For each of the hormonal agents of implementation for regulatory purposes such a particular spectrum of subtle biological as radioimmune assay - with potential sensitivi- effects may occur in various species and strains ties in the part per trillion range, the question of test animals. These effects may vary from emerged what level of sensitivity should be one animal type to another with respect to required in the future. Additional information qualitative differences and dose-response relarelated to certain environmental agents such as tionships. To concentrate on just one measureaflatoxin with known abilities to induce cancer ment of a particular parameter of effect may be when fed at low levels in the diets of experi- to overlook other effects of subtle but substanmental animals presented the more general tial biological importance. Is the biological question: on what basis should such decisions effect noted in one type of test animal directly be made with respect to required methodolo- or indirectly relevant to the human situation? Is gies to detect added substances? The legal this relevance qualitative but at variance with requirement that like things should be treated human dose-response relationships? Is there a in like manner leads to the question, like what? different set of considerations involved when Which criteria should be used for making considering short term higher dose human determinations of required sensitivity? What sort exposures involved with therapy in clinical of an objective reference point could be set up medicine practiced on patients with particular for such purposes that would have a certain characteristics as compared to lower level degree of understanding and support through- longer-term exposures to a more generalized out the scientific community? population? Are certain effects albeit subtle on As you know, FDA proposed in the Federal receptor sites in certain tissues and organs of Register of July 19, 1973 that the Mantel- epidemiological significance with respect to Bryan concept be used to determine what a considerations of reproductive or carcinogenesis conservative safe dose would be with respect to problems? Are effects on animal receptor sites extrapolations to man from data involving proportionate to responses in humans?" Which experimental animals. In turn this virtually safe effects may be additive with respect to other dose would be used to define what sensitivity expected exposures? Conversely, what interwas to be required of the analytical methods ferences might be expected? Is there evidence used to assure the absence of residues. In this to suspect an interaction with other substances manner, it was hoped that objective criteria considered to have carcinogenetic potential? would take into account the dose response What about the unmasking of viral oncogenes? toxicity data pertaining to induction of cancer Is susceptibility to independently caused canin test animals by specific compounds and thus cers increased or decreased by low level exafford a standardized way of relating the posures that may occasionally occur? required analytical sensitivity to the strength of Dr. Friedman was a proponent of the cona particular carcinogenetic agent. cept that a dose-response relationship must Our knowledge about the mechanisms by exist on a physiological basis, and that statisti- Downloaded from jas.fass.org by on January 3, 2010. 1260 KOLBYE protect public health in situations where the benefit to the public may be much more indirect than the direct clinical benefits involved in the practice of medicine. What may well be an acceptable risk in the context of clinical medicine may well be unacceptable with respect to food exposures. And herein lies the dilemma that we are faced with. Until such time as there is clear and convincing scientific proof with respect to the incidence and avoidance of biological thresholds w i t h respect to carcinogenesis, these people who adopt the position that there is no such thing as a safe dose to a carcinogen are in the position of raising questions difficult to answer. Proponents of the multistage mechanism theory involved with the carcinogenesis of particular substances have to meet a heavy burden of proof in order to support a request for exceptions from the more general practices of toxicology dealing with cancer induction. In the meantime, determinations of what would constitute safe usage of hormonal agents in food producing animals must continue. In that respect it would appear that the adoption of some approach such as that of Mantel-Bryan would provide a more predictable and defensible way of setting the required sensitivity standards to assure that no residues of a carcinogen occur in food products of animal origin as determined by a method of analysis approved by the Secretary. The Mantel-Bryan or similar approach to extrapolating risk from experimental data on animals to humans appears to provide a more standardized approach to evaluating the strength of a carcinogen and determining what the acceptable sensitivity should be. Where do we stand now and what considerations are being examined at the present time? Within FDA we have several activities underway. You have heard about the nature of some of the comments received by FDA by other speakers at a variety of meetings. Involved in some of the comments to the FR notice were considerations ranging from the adequacy of the Mantel-Bryan procedure to questions involving specific applications. Currently within the agency resources of both the Bureau of Foods and the Bureau of Veterinary Medicine are involved in various activities designed to elucidate criteria for approaching the problems cal concepts describing the Beta error problem as it applies to toxicological testing for carcinogenetic potential were helpful but not fully descriptive of the phenomena involved in the biological responses of test animals and humans. His approach was that future research would likely elucidate multistage mechanisms involved in the actions of many carcinogenic substances. I tend to share his opinion as do many others. Certainly it is logical to suspect that whatever changes occur to trigger malignant cellular growth are grounded on more fundamental alterations in basic physiology. We know that such is generally true for most other considerations involved in toxicology as applied to food safety determinations. If in fact a biological threshold exists before a subsequent stage immediately preceding malignancy occurs, and such threshold is not attained, it certainly is logical to assume that the second stage will be avoided altogether and the occurrence of malignancy prevented. Titrations are frequently used in clinical medicine to maximize therapeutic effects of a drug consistent with minimizing adverse side-effects. And certainly such an approach is taken with regard to using hormonal agents directly in humans. With respect to some parameters of toxicity of various hormonally active agents, we have been examining the problems of species and strain specificity of biological effects in relation to dose in various test animals. We are attempt~ g to assess their relevance to the human situation. For certain substances we have suggested that further testing be performed in larger anihaals such as monkeys in order to gain a greater appreciation of the relative biological strength of the compound in question. This information may be very useful in "calibrating" the response of smaller test animals in relation to humans. One substance under FDA supported study is zearalenone because of potential implications in relation to food safety if weather conditions adversely affect crops to permit certain fungal growths. Similar information may be desirable for certain hormonal agents intended for use in food producing animals. In dealing with questions of food safety, the Bureau of Foods must take into consideration the current state of the scientific arts of toxicology and oncology and must serve to Downloaded from jas.fass.org by on January 3, 2010. ANTIMICROBIALS AND HORMONES FOR LIVESTOCK 1261 mentioned earlier in this talk. tion? The no residue proposal separated the Experimental data are being examined by a endogenous hormones from exogenous drugs computerized program in order to get some fix and described the requirements for assuring on where we stand with certain agents including that background levels would not be raised in hormonal ones that have already been approved such a way that expected human exposure to for use. Questions abound as to what particular such agents would increase. approach should be used in applying the model More research is indicated on the endoto particular data. What should be included in genous hormones and their metabolites to the numerator and what should be included in determine various characteristics of importance the denominator at what point in time in order to food safety determinations. Analytical methto obtain dose-response data points for extra- ods need to be improved in order to detect polation? Should male and female animals be residues occurring in tissues of recipient animals treated as a pool or in dealing with specific to assure that serum measurements will acsex-related tumors should differentiation be curately reflect a known pattern of distribution made? What effect would the single hit hypo- within the body's metabolic pool of hormonalthesis have on "safe" doses at various para- ly active parent compound and metabolites meters of risk? What other factors influence with significant hormonal activity. Only in this risk and how do we take them into account? way will the addition of naturally present The experimental data usually involve lifetime hormones to the animals be capable of being feeding studies where a general level of the monitored to assure that the average levels of substance in question has been added on a daily endogenous hormonal substances in food of basis to the diet of the control animals. If there animal origin remain within normal limits. is a low probabilitY that a particular food Potential problems are involved when one commodity might contain residues below the attempts to define so-called normal limits; limits of detection, to what degree is that much careful survey work needs to be done. particular food commodity important in the Exogenous substances and related metabodiet? How can we account for variation in lites were defined as those not naturally present dietary habits in order to protect those people in the recipient animals. For these the classical who consume above-average amounts of the testing procedures to detect adverse effects on commodity in their diet? reproduction and potential carcinogenic effect Another question enters into consideration: have been required. If carcinogenic, then the how should we take into account those agents proposed Mantel-Bryan procedure would apply which are endogenous to the food producing and the so-called virtually safe dose levels could animal such as naturally occurring hormones? be determined and translated into the required Quite possibly, data on carcinogenesis experi- analytical sensitivities of the test methods. ments in some instances could lead to situations where the "safe" no residue levels by Mantel- The comments made by Dr. R. G. Zimbelman as a Bryan techniques were lower than the naturally discussant are essentially contained in a paper entitled "Approach to Hormones as Drugs for Animals - Relaoccurring levels of the same compound in food tionship to Delaney amendment" which was published producing animals and their products. How can in the Journal of Animal Science, Supplement 1, such background levels be taken into considera- 38:68. George H. Gass, ~ Discussant Dr. Preston has summarized the situation of hormonal additives in foodstuffs extremely well. One might take exception to even the use 1Physiology department, Southern Illinois University, Carbondale, Illinois 62901. of physiologically active exogenous hormones (whether natural or synthetic) that are carcinogenic, in dosages above the normal physiologically active range. A curve showing one of the physiological e f f e c t s o f estrogens shows them to be similar as regards their effect on uterine Downloaded from jas.fass.org by on January 3, 2010.