![[logo] US EPA](../gif/logo_epaseal.gif){kind=logo}
Final Report
A Set of Scientific Issues Being Considered by the Agency in Connection with the Criteria for Requiring In-Utero Cancer Studies
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific Advisory Panel (SAP) has completed its review of the set of scientific issues being considered by the Agency in connection with the criteria for requiring in-utero cancer studies. The review was conducted in an open meeting held in Arlington, Virginia, on September 9, 1997. The meeting was chaired by Dr. Ernest E. McConnell. Other Panel Members present were Dr. John Adgate (University of Minnesota School of Public Health); Dr. Charles Capen (Ohio State University); Dr. Robert Chapin (NIEHS); Dr. Samuel Cohen (University of Nebraska Medical Center); Dr. Terri Damstra (Tulane University); Dr. Ronald Kendall (The Institute of Environmental and Human Health, Texas Tech University/Texas Tech University Health Sciences Center); Dr. Genevieve Matanoski (Johns Hopkins University); Dr. Michele Medinsky (Chemical Industry Institute of Toxicology); Dr. Harihara Mehendale (Northeast Louisiana University School of Pharmacy); Dr. Lorenz Rhomberg (Harvard Center for Risk Analysis); Dr. Mary Anna Thrall (Colorado State University); and Dr. John Wargo (Yale University).
Public Notice of the meeting was published in the Federal Register on July 14, 1997.
Oral statements were received from the following:
Dr. David Wallinga, National Resources Defense Council
Mr. Richard Wiles, Environmental Working Group
Written statements were received from the following:
American Crop Protection Association
QUESTIONS FOR SAP
1. Please comment on the set of proposed factors to be considered in determining the need for perinatal carcinogenicity testing of a pesticide.
a) Should other factors be considered?
b) Should any of the proposed factors be dropped or modified?
RESPONSE:
The SAP compliments the Agency for responding effectively and appropriately to the suggestions of the previous SAP review of the Agency's proposed guidelines for perinatal carcinogenicity testing. In response to the SAP's suggestions the Agency reviewed the available information, including the FDA guidelines, and developed a set of factors appropriate for application to this issue.
The SAP agrees with the Agency's conclusion that perinatal carcinogenicity testing is not appropriate on a routine basis. The consensus of the SAP is that this testing would not usually contribute significantly to the qualitative determination of carcinogenicity presently assessed using adult animals. One recommendation is that the EPA pursue this issue more broadly, within a research, rather than testing, context.
a) The Panel consensus was that the list of factors proposed for determining the need for perinatal carcinogenicity testing was appropriate and that no other factors needed to be considered. There was some discussion among Panel Members regarding the concept of weighting or prioritizing these factors. The SAP agrees with the general concept of weighting but did not want to prescribe any one formula at this point in the development of the policy. The majority of Panel Members favored the position that,in order to give the agency maximum flexibility in these initial stages of policy development, the proposed approach of giving equal weight to all factors was the most beneficial. The Panel considered it likely that in practice, depending upon the pesticide under consideration, one factor might be weighted more heavily than another and that critical factors might change for various pesticides under consideration.
For example factors d, e, and f, which relate to induction of DNA adduct formation and cell proliferation in fetal/perinatal tissues, may not of themselves be of sufficient weight to require perinatal testing. Increases in some factors may be accompanied by decreases in others. In these cases additional testing may not be required. As an example, DNA adduct formation in fetal tissue might be accompanied by destruction of initiated/damaged cells by apoptosis. Observations of increased rates in these processes do not necessarily translate to increased cancer incidence. For example in adult carcinogenesis studies, diet restriction stimulates cell proliferation, but this response is accompanied by an equally stimulated rate of apoptosis. As a result diet restriction decreases cancer incidence.
b) The SAP was of the opinion that none of the factors should be eliminated or modified. The consensus view was that depending upon the database for the pesticide under consideration one factor might be considered more significant than others.
2. The Office proposes to use a weight-of-the-evidence approach in applying the proposed factors in determining if a pesticide should be a candidate for perinatal carcinogenicity testing.
RESPONSE:
The Panel approved of the weight-of-the-evidence approach. The proposed policy and the factors to be considered appear to be reasonable but have been developed in the absence of specific examples. Thus, the Panel recommends that the process be tested using five to eight specific chemicals as examples to determine the usefulness of the policy in practice. The Panel requests that the Agency bring the conclusions and supporting evidence to the SAP or another peer-review group for additional discussion.
The SAP applauds the goal of the Agency to be as clear as possible about those factors to be applied in the decision to trigger a perinatal carcinogenicity test, although the SAP realizes that it is not possible, nor desirable, to prescribe specific factors or weights at this time.
In the absence of a clearly emerging consensus on how to weigh the individual factors, the Agency should study the value of identifying a threshold weight-of-the-evidence as a trigger point to require a perinatal carcinogenesis study. This method may be used until such time as when more definitive consensus is developed on weighing each factor.
The Panel gave some attention to the issue of the direction in which the factors should be applied. The SAP favors the approach that the weight-of-the-evidence be used coming from the presumption of "no need to conduct" the test, with the various factors being evaluated in order to change the presumption from "not testing" to "testing". In other words there must be clear justifications to test rather than not test. However, the Panel also recognizes that an alternate approach is the reverse position that is a presumption that the testing is needed with the various factors applied to obviate the requirement for testing.
FOR THE CHAIRPERSON:
Certified as an accurate report of findings:
Larry C. Dorsey
Designated Federal Official
FIFRA/Scientific Advisory Panel
DATE:
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific Advisory Panel (SAP) has completed its review of the a Set of Scientific Issues Being Considered by the Agency in Connection with the Exposure Assessment Methodologies for Residential Scenarios. The review was conducted in an open meeting held in Arlington, Virginia, on September 9, 1997. The meeting was chaired by Dr. Ernest E. McConnell. Other panel members present were Dr. John Adgate (University of Minnesota School of Public Health); Dr. Julian Andelman (University of Pittsburgh); Dr. Charles Capen (Ohio State University); Dr. Robert Chapin (NIEHS); Dr. Samuel Cohen (University of Nebraska Medical Center); Dr. Ronald Kendall (The Institute of Environmental and Human Health, Texas Tech University/ Texas Tech University Health Sciences Center); Dr. Genevieve Matanoski (Johns Hopkins University); Dr. Michele Medinsky (Chemical Industry Institute of Toxicology); Dr. Harihara Mehendale (Northeast Louisiana University School of Pharmacy); Dr. Lorenz Rhomberg (Harvard Center for Risk Analysis); Dr. Mary Anna Thrall (Colorado State University); and Dr. John Wargo (Yale University).
Public Notice of the meeting was published in the Federal Register on July 14, 1997.
Oral statements were received from the following:
Dr. Jeffrey H. Driver, Pyrethroid Working Group
Ms. Laura Fell, Johnson Wax - representing Chemical Specialties
Manufacturers Association
Dr. David Karmol, National Spa & Pool Institute
Mr. Brad Shurdut, DowElanco - representing American Crop Protection
Association
Written statements were received from the following:
Chemical Specialties Manufacturers Association
Pyrethroid Working Group
California Environmental Protection Agency
Subject: Agency Questions for the SAP on the Residential Exposure Standard Operating Procedures (SOPs)
1. Do residential exposure scenarios included in the SAP background document represent all significant exposure scenarios? If not, what other scenarios should be included?
RESPONSE:
a. The individual scenarios seem to be appropriate, and we have no additions to suggest.
b. There is a common implicit assumption that products will be used or applied at the label or recommended rate or less. It is prudent to assume that some "reasonable misuse" will occur whereby people ignore label instructions and apply more of the product under the assumption that "more is better." While additional scenarios would not have to be developed, the document should acknowledge this possibility and try to quantify overuse by bounding estimates: e.g., the effect of using two cans of a product when one is called for; repeated applications at inappropriately short intervals; or application to inappropriate locales or materials.
c. Little guidance is provided by the Agency to estimate exposure across media.
d. Similarly, little guidance is provided to account for the probability of exposure to mixtures, including mixtures that may act via the same mechanism.
e. The Agency should explicitly recognize the problem posed by inert ingredients, which contribute to the mixture and multi-media problems described in (c) and (d) above.
f. The details of specific computer models used to estimate exposure were not provided to the SAP for its review, and we hope that this will occur in the future.
g. The term "residential" needs clarification. Does it include neighborhood pools at condominia, apartment complexes, and gated residential communities? How should the possibility of contamination originating on nearby non-residential properties be considered in exposure estimates (e.g., spray drift from agricultural applications adjacent to a residential community?)
h. The Agency should provide additional guidance regarding the need to consider the distribution of pesticide contamination across space and time within residential environments. This might include probable room-to-room variance and estimates of chemical persistence. Also, how might exposures accumulate across time?
i. The EPA document should be more explicit as to its exact purpose and use, indicating, for example, whether it will be used primarily as a screening document for regulators making decisions about approval of pesticide formulations or used in the actual assessments of exposure and risk following pesticide use. This information is essential to enable reviewers to properly critique this document or subsequent versions.
2. Is the document sufficiently explicit in explaining the source and rationale for assumptions and standard factors used for calculations: If not, where should further documentation or discussion be added?
RESPONSE:
a. No. The Agency should provide considerable additional rationale for suggesting default assumptions to be used in the absence of credible data. Appendix A contains many different statistical summaries proposed as default assumptions. Without explanation, mean values are suggested for some scenarios while 90th-percentile levels are proposed for others. If summary statistics are proposed, consistent principles--as yet not articulated--should justify their selection.
b. The Agency has not yet considered the effect of combining numerous conservative assumptions when using statistical summaries. Although it is commonly assumed that this method will produce gross overestimates of exposure, it is also possible that exposure will be significantly underestimated. This can occur when 90th-percentile values of contamination or intake distributions are commonly zero, i.e., when less than 10% of samples were contaminated. Under these conditions the entire exposure estimate will be driven to zero, while the original selection of the 90th percentile may be intended to be conservative. When distribution of contamination and intake are more normally distributed, the use of summary statistics will quickly build conservatism in exposure estimates. Unfortunately, it is not easy to predict the shape of these distributions, given the general absence of chemical-specific contamination and intake data. Thus, the use of summary statistics to represent default assumptions will commonly result in significant errors in exposure estimates.
c. The problem of compounding conservatism may be avoided using probabilistic methods. They provide a more accurate image of the likelihood of exposure at different levels. Generally, we conclude that it would be most desirable for the Agency to estimate exposure probabilistically, if credible distributions of data exist.
d. This suggestion to estimate exposure probabilistically creates additional questions. When are the data of sufficient quality to justify their use in exposure assessment, as opposed to relying upon default assumptions? This is the central question of interest to pesticide registrants who will encourage the Agency to use less conservative assumptions.
e. In general, it was difficult for the SAP to understand the rationale and scientific support for activity and behavior assumptions. How should behavior variance be incorporated into exposure estimates in a systematic and realistic manner? Infant hand-to-mouth behavior may be a significant route of infant exposure to residues that persist on floors, carpets, furniture and toys.
f. The intended level of protection remains unclear. For example, if one assumes that protection of 95% of the population of 2 year-olds is adequate, this will leave 5%--or 200,000 children of that age at higher--risk in the U.S.
g. Generally, additional documentation is necessary to support the choice of exposure scenarios, models, data sets and default assumptions to be used in the absence of adequate data. For example, each equation should be supported by current and credible scientific literature. The SAP expects that the Agency will continually break new ground in the area of exposure assessment, however, it would be helpful for us to understand when the Agency believes this is occurring.
h. The algorithms for assessing inhalation exposures from pesticides in swimming pools appear to be flawed and are likely to substantially overestimate such exposures.
3. Are the example calculations provided in the background document clear and helpful to the reader? If not, what modifications would make them more useful?
RESPONSE:
a. The example calculations are very useful and should be more numerous.
b. They could be expanded to demonstrate the difficulty of estimating exposure when data quality varies among components of exposure equations.
c. For example, they might be developed to demonstrate probabilistic methods in contrast to simpler combinations of statistical summaries.
d. Or they might be developed to consider the sensitivity of exposure estimates to spatial and temporal variance in contamination, or behavioral variance among those potentially exposed.
e. Examples should be developed to consider the problem of estimating exposures to mixtures that might act via the same mechanism.
f. Similarly, examples might be developed to demonstrate the problem of estimating the accumulation of exposures across different contaminated media and across time.
4. Are the limitations and uncertainties associated with these scenarios discussed in sufficient detail? If not, what additional information should be included?
RESPONSE:
a. The SOP's should consider the different sources of uncertainty associated with each component of the exposure estimates--normally contamination estimates, and estimates of how individuals will encounter contaminants via different routes. For example, uncertainty in a contamination estimate might have many sources, including an inappropriate sampling design (failure to capture the distribution across space, time or demographic characteristics); aged data no longer reflecting current conditions; or misunderstanding chemical persistence, its movement through the environment, and/or its fate.
b. Clearer understanding of the sources and magnitude of uncertainty should have many effects. First, it should clarify how uncertainty is compounded as variables are combined in exposure equations. Second, it should help the Agency to make difficult judgments regarding which sources of uncertainty are most important to protect those believed most susceptible to adverse health effects.
c. The 1996 Food Quality Protection Act requires that EPA consider all possible sources of exposure when making registration and tolerance decisions. Further, it requires that the potential for exposure to mixtures of chemicals be formally considered if they appear to operate via the same mechanism of toxicity. To meet these requirements, EPA should develop principles for managing uncertainty--uncertainty that grows exponentially as methods are employed to estimate exposure to mixtures of pesticides that might be contaminating diverse environmental media.
5. Are the scenario SOP's a reasonable approach for conducting a first tier, screening level residential exposure assessment when chemical/product specific-data are not available? If not, which SOP's and what alternative approaches/data sources are recommended?
RESPONSE:
a. The general approach suggested seems very reasonable and appropriate given the inadequacies of available data.
b. The dominant limitation is the over reliance of the approach on assessment of a single medium of contamination and exposure.
c. The Agency should quickly develop methods to account for exposure to mixtures of chemicals (acting via the same mechanism) across diverse environmental media.
d. In the absence of reliable data needed to produce credible exposure estimates, the Agency's proposed reliance upon conservative default assumptions about levels of exposure appears to be justified.
6. Since these residential SOP's are intended as a Tier 1 approach and are based on various assumptions, not chemical specific data, is the use of the additional tenfold margin of safety appropriate? FQPA (Section 408 (b) (2) (C) (ii) (II) states that "in the case of threshold effects,...an additional tenfold margin of safety for the pesticide chemical residue and other sources of exposure shall be applied for infants and children to take into account...completeness of the data with respect to exposure...to infants and children.")
RESPONSE:
It is not possible to judge if it is appropriate to incorporate an additional tenfold safety factor for specific exposure scenarios given the unknown magnitude of the uncertainties and conservative nature of some assumptions.
a. The entire approach being proposed by the Agency is a further refinement of the requirement by FQPA that a tenfold safety factor be employed to manage error in exposure estimates. In other words, it appears that the Agency is already applying a safety factor concept within the default assumptions.
b. If the Agency employs only summary statistics to define default assumptions, it will be difficult to predict if a tenfold margin of safety remains for reasons cited in 2(b) above.
c. There appear to be different types of default assumptions relevant to choosing a safety factor. First, what should be assumed in the absence of data necessary to make credible and reasonably accurate predictions of exposure? Second, what methodological assumptions should be made about combining estimates of exposure across scenarios and across mixtures that may contribute to the same adverse health effect? Thus, data sufficiency alone does not obviate the need for default methodological assumptions.
d. If chemical-specific data (deemed to be of sufficient quality) are employed, expert judgment concerning their quality will be necessary to determine if an additional ten-fold safety factor is warranted. Again, it is certainly plausible that the Agency could incorporate safety factors explicitly within default methodological assumptions in a manner that would obviate the need for an additional ten-fold level of protection. It is not yet clear how this would be accomplished. The changes suggested here would help the Agency meet the letter, intent, and spirit of the FQPA language.
FOR THE CHAIRPERSON:
Certified as an accurate report of findings:
Larry C. Dorsey
Designated Federal Official
FIFRA/Scientific Advisory Panel
DATE:
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific Advisory Panel (SAP) has completed its review of the efficacy testing issues concerning public health antimicrobial pesticides. The review was conducted in an open meeting held in Arlington, Virginia, on September 10, 1997. The meeting was chaired by Dr. Ernest E. McConnell. Other panel members present were: Dr. Charles Capen (Ohio State University); Dr. Rick DiGiulio (Duke University School of the Environment); Dr. Michael Doyle (University of Georgia); Dr. Larry Duda (Pennsylvania State University); Dr. Elwood Hill (Independent Consultant-Gardnerville, NV); Dr. Ronald Kendall (The Institute of Environment and Human Health, Texas Tech University/Texas Tech University Health Sciences Center); Dr. Michele Medinsky (Chemical Industry Institute of Toxicology); Dr. Harihara Mehendale (Northeast Louisiana University School of Pharmacy); Dr. William Rutala (University of North Carolina School of Medicine); Dr. Dhiren Shah (Food and Drug Administration); Dr. James Shih (National Institutes of Health); Dr. Lynne Sehulster (Centers for Disease Control & Prevention); Dr. Mark Sobsey (University of North Carolina); Dr. Mary Anna Thrall (Colorado State University); and Dr. John Wargo (Yale University).
Public Notice of the meeting was published in the Federal Register on July 14, 1997.
Oral statements were received from the following:
Mr. Richard Dimock, Safe Tables our Priority (STOP)
Dr. Joseph F. Frank, University of Georgia
Dr. Conrad J. Heilman, Jr. Quality Biotech
Dr. Robert Kiefer, Chemical Specialties Manufacturers Association
Dr. Patricia Marion, Stanford University
Mr. Niranjan C. Patel, Gibraltar Laboratories
Ms. Joan Popowics, Proctor & Gamble
Ms. Susam M. Price, 3M Corporation
Mr. Robert G. Prince, Gibraltar Corporation
Mr. James C. Wright, representing Gibraltar Laboratories
Written statements were received from the following:
3M Corporation
Dr. Larry Beuchat, University of Georgia
Dr. Linda Harris, University of California-Davis Lonza, Inc.
Mr. James C. Wright, representing Gibraltar Laboratories
A. QUESTION ON CRITERIA AND SYSTEMATIC PROCESS FOR ACCEPTANCE OF NEW OR MODIFIED EFFICACY TEST METHODOLOGY
(1) The Agency maintains the scientific position that use of standardized and validated antimicrobial efficacy test methodology is essential for validation, comparison, and enforcement of antimicrobial product effectiveness. The Agency proposes to establish specific written criteria that new methods or modified existing methods must meet to be considered valid. Included in such criteria is the requirement that new and modified methods be subjected to laboratory validation processes prior to EPA acceptance. Does the Scientific Advisory Panel concur with proposed criteria and laboratory validation processes, or recommend an alternative approach?
RESPONSE:
New or modified methods for determining the antimicrobial efficacy of products should meet the minimum performance standards defined for the "standard procedure." Ideally, these new methods should be validated through the AOAC Official Methods Program or an equivalent methods verification program; however, the time for approval of such methods can unduly delay approval of products of major importance to public health. Given the rapid changes in technology, EPA needs to maintain open-mindedness and flexibility in allowing new and improved test methods to be used, with the proviso that they be validated, reproducible, accurate, and precise. While the Panel appreciates the rigor of the AOAC procedure, it would be preferable to have a more streamlined process. Hence, alternate approaches should be considered for products.
The AOAC Peer-Verified Methods Program that would use four to five peer reviewers, rather than two, would be an acceptable alternative. For modified efficacy methods, validation by either the AOAC Peer-Verified Methods Program (or an equivalent method) or by the AOAC Official Methods Program would be appropriate.
The Panel suggests that the Agency reviews its choice of a "gold standard" periodically, e. g., every three years. Other organizations, such as ISO and ICMSF, could perhaps be considered as resources.
The efficacy test utilizing nutrient rich media and the current environmental conditions for growth of microbes is outdated for the following reasons:
a. Microbes present on inanimate surfaces or living tissues usually are not present as planktonic cells. They attach to surfaces, may form biofilms, and may produce exopolymers that may retard or prevent penetration of antimicrobials and sanitizers. Microbes present in the natural environment may have induced resistance to environmental insults such as pH, temperature, etc. Their response to antimicrobials could be very different compared to cultures grown in rich media in the laboratory. Development of new methods in this area should take these characteristics of microbes into consideration to make efficacy data realistic for specific applications.
b. Quantitation of microbes following treatment with sanitizers/antimicrobials leads to injury of the surviving population. Injured cells are not recovered on selective media commonly employed in efficacy test protocols and this loss leads to exaggerated efficacy. Use of antibiotic resistant test strains and antibiotic-incorporated nutrient media for recovery and quantitation of survivors should be considered in all method development.
B. QUESTION ON PHENOL RESISTANCE
(1) What scientific direction should be taken regarding the lack of standard and uniform resistance levels to phenol of the test cultures used in the existing AOAC test methods?
--Should the Agency totally eliminate the phenol resistance requirement; or
--Modify the required phenol resistance patterns to provide a broader range of acceptable resistance; or
--Replace the phenol resistance requirements with some other procedures that assure hardiness and equivalence of test cultures, such as standard, quantitative inoculum level.
RESPONSE:
There is no current relevance to requiring the phenol resistance test; hence phenol coefficient method should be eliminated and new protocols should be established for defining the conditions for culturing test microorganisms with suitable resistance levels to antimicrobials. For example, recent studies have revealed that E.coli 0157:H7 strains have induced tolerance to acid and certain environmental stresses. This pathogen is unusually tolerant to acid and has been associated with several outbreaks of infection involving acidic foods (such as apple juice, yogurt, and fermented sausage) that were previously considered safe from foodborne pathogens. These data indicate that E.coli 0157:H7 should be one of the principal test organisms for evaluating the efficacy of antimicrobials and sanitizers. The methods used to culture E.coli 0157:H7 to induce acid tolerance need to be critically defined, and phenol resistance would not be one of the conditions. Induced acid tolerance has been demonstrated in other nonpathogenic E.coli strains, as well as Salmonella species. The regulatory mechanisms of induction to acid tolerance are similar in these organisms and it is most likely that they are global.
In summary, phenol resistance does not appear to be linked to other germicidal or antibiotic resistance nor is its resistance a good marker of genetic variability (or similarity). Furthermore, levels of phenol susceptibility frequently are outside current EPA guidelines, resulting in significant expense to industry and government.
Quantitative inoculum level is a better choice, if growth and environmental conditions suggested for any new methods are considered. However, current methodologies for quantitation are not accurate, as they do not take into account inadequate removal of microorganisms from natural substrates and the ability of the organisms to become more resistant to pH change, oxidation, altered temperature, etc. Research is needed in this area. The Panel recommends that the Agency request a consensus approach of available experts who can define reproducible tests which model germicidal resistance in nature.
C. QUESTIONS ON TREATED ARTICLES
(1) If the Agency were to consider registration of treated articles with public health claims, especially for food contact surfaces, are the existing test methodologies and efficacy performance standards appropriate to determine the efficacy of such articles against pathogenic organisms?
RESPONSE:
The present approach of simulated in-use test methodology is appropriate, and the five Agency recommended related elements are correct and inclusive. However, some methodologies and performance standards should be reevaluated and updated. For example, demanding 3-log reduction of organisms may not be appropriate. In some circumstances, it is better to kill or control pathogens with an antimicrobial chemical (one that may not be as efficacious as a sanitizer but capable of killing 100-fold of the target pathogens) than to have no treatment available. The time frame necessary for antimicrobial agents to be effective may also need to be modified. A new category for public health claims should be considered, such as "antimicrobial treatment" or simply "antimicrobial", that would kill 2-log (100 fold) of the target pathogens per ml or gram. The target pathogens should include those that are significant foodborne pathogens that have been identified as public health problems associated with the articles to be treated. Appropriate bacteria would likely be E.coli 0157:H7 (representing gram-negative bacteria) and Listeria monocytogenes (representing gram-positive bacteria). Five strains of each which were previously associated with foodborne outbreaks should be used.
Incorporating a bacteriostatic germicide into a sponge to prevent growth and dissemination that could be used in food preparation areas is an example of a product that could possibly enhance health. A recent study reports that 49% of Americans use a sponge or dishtowel to clean areas which contain meat and poultry juices, and such juices may contain pathogens such as Salmonella, Listeria monocytogenes and Camplyobacter jejuni. These sponges and cloths may re-deposit pathogens on other surfaces since bacteria can grow to populations of more than one billion in a sponge within 24 hours of use.
The idea of including a germicide in sponges is an interesting concept. However, current methodology used by some companies to evaluate the efficacy of treated sponges is likely inadequate. Bacteria tested should include five outbreak-associated strains each of Salmonella, Listeria monocytogenes, and E.coli 0157:H7. Also, specific available tests for measuring efficacy must be reconsidered because of complications associated with microbes forming biofilms on surfaces, penetration of microbes into the article, and adhesion of the microbe films. The Agency should also consider if health safety issues will result from antimicrobial agents acting as indirect food additives.
(2) For most treated articles, the presence of moisture is a key factor in activating the antimicrobial agent against microorganisms. For those situations where the surfaces are routinely dry and the antimicrobial agent releases slowly, are there acceptable or recommended methodology(ies) and performance standards for determining efficacy of public health claims?
RESPONSE:
Migration of antimicrobials for some synthetic articles is not dependant on moisture. The relative humidity and moisture have to be included in establishing efficacy data.
(3) Is the state-of-the-science such that the Agency needs to develop new test methodologies and/or efficacy performance standards for treated articles with public health claims?
RESPONSE:
Yes. New test methodologies need to be developed. New methods of removing bacteria from surfaces should be developed because of penetration of organisms under surfaces as well as development of "biofilms". Moreover, methods of quantitating antimicrobial diffusion should be established.
D. QUESTIONS ON FRESH FRUIT AND PRODUCE SANITIZING WASH
(1) Since this is a new and unique area for the Agency, are the current testing methodologies and performance standards sufficient to determine the efficacy of products which are used to sanitize fresh produce (fruits/vegetables) at the homeowner (household) level?
(2) If the current procedures are not appropriate, then what efficacy methodology(ies) and performance standards would need to be developed to ensure acceptable sanitization by the consumer?
(3) If further research is needed to develop new methodologies and standards, are there interim methodology(ies) and performance standards that the Agency can utilize in assessing the efficacy of such sanitization techniques?
RESPONSE:
The best methods available for removing or killing pathogens in fresh produce are likely inadequate to assure public health. A water wash followed by a rinse with up to 200 ppm hypochlorite solution reduces pathogen populations by approximately 10 to 100-fold. Considering the concerns that potential carcinogens are formed in water and food from hypochlorites, the use of such treatments is inadvisable. Presently, consumers have very little available to them to treat produce in order to substantially reduce the microbial load. Hence, treatments that kill or reduce microbial loads on produce by 100 fold or more would be significant. In the interest of public health, treatments in this category, which could be registered as an "antimicrobial wash," should be approved for use as an interim measure. A 2-log reduction of five outbreak-related strains each of Salmonella, Listeria monocytogenes, and E.coli 0157:H7 could be a reasonable performance standard because, when pathogens are present on fresh produce, they are usually present at populations of less than 100/gm.
The National Advisory Committee on Microbiological Criteria for Foods Subcommittee on Produce (NACMCF) has been addressing the very difficult issue of enhancing the microbiological safety of fresh produce and products of fresh produce. The Subcommittee is comprised of many food safety experts from public health agencies (including CDC), federal regulatory agencies (including FDA and USDA-FSIS) and academia (including the University of Georgia Center for Food Safety and Quality Enhancement). This group has invested considerable time and effort in investigating the problems and difficulties associated with the microbiological safety of produce. The EPA is encouraged to take advantage of the NACMCF's expertise and include the Subcommittee's input in developing testing methodologies and performance standards. In summary, the EPA should move to facilitate the development of validated standard techniques for determining the efficacy of germicides to be used on fresh fruits and vegetables.
E. QUESTIONS ON EFFICACY TESTING FOR HEPATITIS B-TYPE VIRUS
(1) If the Agency decides to replace the chimpanzee test used in testing the efficacy of disinfectants against human hepatitis B-type virus, what test methodologies could be used as a replacement. Two possibilities that have been proposed to the Agency are the duck hepatitis B Virus Test (DHVT) and the Morphological Alteration and Disintegration Test (MADT). Could one or both of these tests, or additional tests be used to test for efficacy against Human Hepatitis Virus B.
RESPONSE:
The Agency has acknowledged the growing concern about the continued use of chimpanzees, an endangered species of primates, as the accepted animal model in liquid chemical germicide efficacy studies with respect to hepatitis B virus (HBV). The Panel concurred with the notion that it is unethical to continue to require such testing under these circumstances. There is a long history of using surrogate microbes to assess the efficacy of disinfection/sterilization technologies against various classes of microorganisms.
Surrogate viruses for HBV exist within the same virus family (i.e., Hepadnaviridae), and other animal models may be used to examine the effects of liquid chemical germicides on such viruses. One candidate for that of an HBV-surrogate virus is the duck hepatitis B virus (DHBV). DHBV has many of its biophysical and biochemical properties in common with those of HBV, and it is very easy to work with in the laboratory. Additionally, the effects of chemical germicides on DHBV can be examined in both in vivo and in vitro settings, thus allowing a full evaluation of viral inactivation subsequent to liquid chemical germicide treatment. Ready availability of the animal host for DHBV also allows for sufficient numbers of animals to be used for each set of experimental conditions so that statistically significant results can be obtained. The question was raised regarding the level of the DHBV's resistance to the action of liquid chemical germicides and how it compares to that for HBV. The Panel discussed the possibility that DHBV may be more resistant than HBV to germicidal chemical activity but, in essence, felt that this was not a serious issue, given that the pathogen of concern, HBV, has been shown to be sensitive to the activity of a wide spectrum of liquid chemical germicides including low-level disinfectants.
Additionally, the Panel acknowledged that it is considered standard to test selected microorganisms from each of the various major microbial classes with chemical germicides and to interpret the results from those tests as representing the general level of inactivation expected for that class of microorganism, despite the possibility that resistance levels to the action of liquid chemical germicides may vary somewhat among the organisms within the class.
Another candidate test considered to replace the chimpanzee test for HBV is the Morphological Alteration and Degradation Test (MADT). The Agency presented basic information about the test protocol and described the advantages and disadvantages of this procedure. Briefly, the test uses human HBV and uses electron microscopy to assess the effect of liquid chemical germicides on the virus in vitro, the indicators of which include viruses which appear to be destroyed and those viruses with morphologically altered surfaces. The Panel did not discuss this procedure at length, although comment was made that the test is subjective. The Panel was in agreement that surrogate viruses and substitute animal models could serve in place of the chimpanzee test for HBV and liquid chemical germicides, but the Panel did not exclude use of the MADT.
(2) If a surrogate test system (i.e., the DHVT) is found to be acceptable for efficacy testing using Hepatitis Virus B, then would the results be sufficient to allow the registrant to make a label claim that the product was efficacious against human hepatitis B virus, even though it was tested against a surrogate virus (i.e., duck hepatitis B Virus) and not the human virus?
RESPONSE:
The Panel discussed this issue only briefly. It was mentioned that given the current requirements for label claims, it was believed that registrants who use the duck virus (DHBV) could make a label claim of product efficacy to either the specific virus or perhaps to the virus family as a whole group i.e., Hepadnaviruses. Other Panelists noted that the Agency has established a precedent where label claims are concerned with Mycobacterium tuberculosis. The Agency has allowed claims for Mycobacterium tuberculosis when registrants have used Mycobacterium bovis as the test organism to represent the Mycobacteria. Agency Staff informed the Panel that the labeling requirements would be revisited to see if this precedent can be extended to the area of HBV efficacy testing. The Panel agreed with this approach. If tests validate that the surrogate virus is less or equally susceptible to inactivation by disinfectants, then any product logically which demonstrates efficacy against the surrogate virus should be allowed a label claim against HBV.
FOR THE CHAIRPERSON:
Certified as an accurate report of findings:
Larry C. Dorsey
Designated Federal Official
FIFRA/Scientific Advisory Panel
DATE:
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific Advisory Panel (SAP) has completed its review of the progress report on developing probabilistic risk assessment methodologies for aquatic and terrestrial risk. The review was conducted in an open meeting held in Arlington, Virginia, on September 10, 1997. The meeting was chaired by Dr. Ernest E. McConnell. Other panel members present were: Dr. Charles Capen (Ohio State University); Dr. Rick DiGiulio (Duke University School of the Environment); Dr. Michael Doyle (University of Georgia); Dr. Larry Duda (Pennsylvania State University); Dr. Elwood Hill (Independent Consultant-Gardnerville, NV); Dr. Terri Damstra (Tulane University); Dr. Ronald Kendall (The Institute of Environment and Human Health, Texas Tech University/ Texas Tech University Health Sciences Center); Dr. Michele Medinsky (Chemical Industry Institute of Toxicology); Dr. Harihara Mehendale (Northeast Louisiana University School of Pharmacy); Dr. William Rutala (University of North Carolina School of Medicine); Dr. Dhiren Shah (Food and Drug Administration); Dr. James Shih (National Institutes of Health); Dr. Lynne Sehulster (Center for Disease Control & Prevention); Dr. Mark Sobsey (University of North Carolina); Dr. Mary Anna Thrall (Colorado State University); Dr. John Wargo (Yale University); and Dr. Don Wauchope (U.S. Department of Agriculture).
Public Notice of the meeting was published in the Federal Register on July 14, 1997.
GENERAL COMMENT:
The Panel addressed the specific questions raised by the Agency and, where necessary, offers additional recommendations for improvement. In consideration of all matters brought out during the meeting and careful review of all documents presented by the Agency, the Panel unanimously submits the following report.
INTRODUCTION & CONCLUSIONS:
The Panel commends the Agency for its proactive response to the FIFRA Scientific Advisory Panel meeting of May 29-31, 1996 in which a primary recommendation was made for the Agency to develop probabilistic assessments of risk and mitigation based upon the sum total of laboratory and field findings. The Agency is correct in that the Panel takes a rather stringent approach to risk assessment. This is because each risk assessment is unique and beset with its own sources of error and data gaps. Therefore, the stringent standards and goals are necessary in order to obtain adequate core data for sound risk assessment. The Panel recognizes the complexity and uncertainty of comprehensive ecological risk assessment. The Agency was reminded that risk assessment can be no better than the process which includes the variable addressed, sophistication of the experimental design, data input, and the adequacy of the analysis. The Agency appears to have accepted this challenge as indicated by the development of the rather ambitious plan presently under consideration.
The Agency is to be specially commended for its outreach effort involving experts from the scientific community (e.g., Society of Environmental Toxicology and Chemistry, American Chemical Society, ASTM, Society of Risk Analysis) rather than reliance on "internal" sources. It is especially important that the Agency follow through on the plan to present risk assessment protocols and data analysis before peer-reviewed national audiences as well as specialty symposia. Excellent examples of the value of this outreach approach include the Aquatic Effects Dialogue Group and Avian Effects Dialogue Group.
The Panel concludes that the planned progress adequately addresses one of the critical issues raised during the above mentioned 1996 Scientific Advisory Panel Meeting. The Panel looks forward to the next progress report during 1998 and especially to implementation of the new tools and processes in 1999.
SAP QUESTIONS FOR PROBABILISTIC RISK ASSESSMENT METHODOLOGIES FOR AQUATIC AND TERRESTRIAL RISK
In May 1996, EPA's Office of Pesticide Programs (OPP) asked the FIFRA Scientific Advisory Panel (SAP) for comment on its ecological risk assessment methods and procedures. While recognizing and generally reaffirming the utility of the current ecological assessment process and methods, the Panel offered a number of suggestions for improving the process. Foremost among the suggestions was that OPP move beyond the present single-point deterministic estimate screening process and develop the tools and methodologies necessary to do probabilistic assessments.
In cooperation with scientists from the American Crop Protection Association, OPP has developed and initiated a process for identifying and/or developing assessment tools and methods that can support probabilistic descriptions of ecological risks from pesticides. This process is described in the background materials prepared for this SAP meeting. (See Process for Improving FIFRA Tools and Methodologies for Evaluating Ecological Risk)
1. Is the process described likely to achieve the results the Panel envisioned?
RESPONSE:
The Panel agrees the approach of the Agency is along the lines intended and has a reasonable likelihood of success. This positive statement must be offered with caution because the stated goals are ambitious given the limited resources available to the Agency which include the possibility of reassignment of the key players. The ECOFRAM dialogue groups comprise an excellent cadre of scientists to deal with this issue. However, heavy reliance on volunteers with other priorities may be problematic.
2. Does the Panel have any suggestions or recommendations in relation to the process that would further help achieve the results the Panel envisioned?
RESPONSE:
To augment efforts of the ECOFRAM groups, the Panel recommends the Agency commit increased workforce to this effort. Also, the existing working groups must include people expert in mathematical modelling of probabilistic risk assessment to interface exposure and effects initiatives for both aquatic and terrestrial methods.
The Panel is concerned that the working groups appears to be organized into traditional "two cultures" of risk assessment -- toxicologists who characterize hazards and environmental modelers who characterize exposures. While this is logical and perhaps necessary, it has in the past hindered the development of risk assessment tools because of the difficulty in combining the types of approaches and data used by these two disciplines. It is essential that a mechanism be in place to ensure continuous communication of the two groups as their deliberations proceed, so that their recommendations and tools will be compatible.
In their outreach to scientific societies the Workgroup should consider making a series of presentations to the IUPAC 9th International Congress of Pesticide Chemistry, to be held in London next August. This meeting is held every four years and is the most important international pesticide science meeting. Presentations would receive international attention and encourage consideration of these tools by regulatory agencies in other countries.
3. Does the charge to the working groups reflect the appropriate direction for developing assessment tools that will better define potential effects to non-target organisms from the use of pesticides?
RESPONSE:
Yes, The steering committee members are an outstanding group that has augmented itself with the remarkable team assembled under the broader ECOFRAM. This contingent is certainly expected to keep the project on track.
4. OPP has recommended that the working groups focus their initial efforts on direct effects in order to accelerate the program's schedule for gaining experience with probabilistic ecological assessments. Could the Panel comment on the usefulness and limitations of focusing OPP's initial efforts in this way?
RESPONSE:
The Panel agrees with the Agency that the focus on direct effects is appropriate for developing probabilistic ecological assessments. This is not to infer that the indirect effects are unimportant. Indirect effects may be pursued as appropriate as the planned step-wise process develops, but practicality must drive the plan to a degree.
5. Could the Panel provide guidance regarding the level of biological organization on which the working groups should focus their efforts in order to develop probabilistic tools that will have reasonable scientific certainty to support Agency decisions in relation to ecological effects?
RESPONSE:
The existing state of the science indicates that individual and population-level effects are reasonable foci of attention during this phase of the project.
FOR THE CHAIRPERSON:
Certified as an accurate report of findings:
Larry C. Dorsey
Designated Federal Official
FIFRA/Scientific Advisory Panel
DATE: