![[logo] US EPA](../gif/logo_epaseal.gif){kind=logo}
A Set of Scientific Issues Being Considered by the Agency to Determine Antimicrobial Issues
Scientific Advisory Panel (SAP) June 1997 Meeting
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific Advisory Panel (SAP) has completed its review of the Office of Pesticide Programs (OPP) policy for determination of antimicrobial issues. The review was conducted in an open meeting held in Arlington, Virginia, on June 3, 1997. The meeting was chaired by Dr. Ernest E. McConnell. Other panel members present were: Dr. Janice E. Chambers (Mississippi State University); James Fairchild (Midwest Science Center); Dr. Richard Fenske (University of Washington); Dr. Robert Herrick (Harvard University); Dr. Paul Kuznesof (U.S. Food and Drug Administration); Dr. Ronald J. Kendall (Texas Tech University/Texas Tech University Health Sciences Center); Dr. Harihara M. Mehendale (Northeast Louisiana University); Dr. Richard M. Parry, Jr. (U.S. Department of Agriculture); Dr. Stephen Saunders (Frito-Lay Corporation); Dr. Lynne Sehulster (Center for Disease Control); Dr. Edward Stein (OSHA-U.S. Department of Labor);Dr. Mary Anna Thrall (Colorado State University); Dr. Donald Wauchope (U.S. Department of Agriculture); Dr. Willis Wheeler (Wheeler Associates).
Public Notice of the meeting was published in the Federal Register on April 22, 1997.
Oral statements were received from:
Dr. Sally Hayes, Chemical Specialties Manufacturers Association
Dr. J. Michael Kelly, Great Lakes Chemical Company
Mr. R. Bruce Jaeger, Stewart Pesticide Registration Associates
Dr. Don Grant, Pesticide Management Regulatory Agency, Health
Canada
Written statements were received from:
Chemical Manufacturers Association
QUESTIONS FOR THE PANEL ON ANTIMICROBIAL ISSUES
A. Toxicology
Question: Traditionally, EPA has required a full battery of toxicity testing for agricultural pesticides which result in residues on raw agricultural commodities. For non-food and sanitizing uses of antimicrobial pesticides, the toxicology data requirements are proposed as a tiered testing scheme. This approach is similar, but not equivalent, to the FDA's approach. For certain use categories where significant human exposure is expected to occur (swimming pools, aquatic outdoor, human drinking water, animal drinking water), a full food use toxicology data set is required. Is the tiered approach an acceptable approach and does it provide for pertinent scientific data for each tier?
The Agency's efforts to streamline toxicology testing of antimicrobials and sanitizing agents through a Tiered approach is commendable. Based on use pattern, antimicrobial pesticides are grouped into 12 categories. This should facilitate management of regulatory issues in close alignment with their uses and anticipated human exposures. The Tiered approach developed by the Agency is a reasonable approach and the Panel supports the Agency in this regard. The limited testing used to support registration of sanitizers and related products with minimal potential for human exposure can be scientifically supported and will be discussed in more detail later in this section. The Guidance Document is long and difficult to follow but with additional refinement the document can be improved particularly as regards to scientific citations of current literature. Clarity of presentation and unambiguous trigger points indicating next Tier level of toxicity testing (reproductive and developmental testing, postnatal development, chronic/carcinogenicity, etc.) would substantially improve the present document. The Panel also encourages the Agency to continue dialogue with Canadian counterparts to harmonize, clearly define trigger points, and improve the guidelines. Use of the 'threshold of regulation' concept used by FDA and levels of human exposure to trigger Tier I and II toxicity testing is highly encouraged to minimize unnecessary testing in cases of minimal human exposure.
Tier I testing lists a battery of toxicological studies that would be required for all antimicrobials, which now include components of sanitizer formulations used on food-contact surfaces and sanitizers which may be embedded in food-contact plastics and rubber articles. Progression to a higher Tier (additional testing) would occur if estimated dietary exposure to the sanitizer exceeded 200 ppb. FDA also has used a Tiered Approach. For exposures less than 10 ppb, only an acute oral study (rodent) and literature search for potential carcinogenicity issues has been required. Exposures between 10 ppb and 200 ppb have required a 90 day rodent study, a 90 day non-rodent study, and possibly a multi-generation feeding study with a teratology phase in a rodent, and short term tests for genotoxic potential which inform the need for concern about the carcinogenic potential.
For the specific cases of sanitizers, dietary exposure as determined by FDA has often been lower than 10 ppb. For these cases, The Agency's requirements will significantly increase the data demand and the costs of toxicological testing for sanitizer applications that, according to FDA, lead to insignificant exposure. For dietary exposures below 0.5 ppb (which are sometimes estimated in the case of repeat-use rubber or plastic articles), the FDA, at the request of the applicant, may apply its "Threshold of Regulation Policy" which can result in a letter stating that the food additive is exempt from the need for a regulation. No new toxicological data need be generated for this approach, although a literature update on the substance of interest is required. If FDA is not satisfied with the applicant's package, the applicant is informed of the need for a formal petition and the necessary toxicity studies. This flexible policy has been highly successful in freeing up scarce resources in FDA's technical review groups, has benefitted industry by not requiring a food additive petition with its attendant costs and time delays, and has not compromised public health because of the extremely low exposures and resultant low risk. The Agency should reassess its proposal for toxicology studies specifically with respect to sanitizers to avoid excessive requirements for applications that will result in exposures substantially below 200 ppb based on the current state of knowledge.
Examples of clarification needed in the Guidelines include: the exemptions for oral, dermal, and eye irritation testing where the test compound undergoes phase change from liquid to vapor. This Guideline assumes that exposure to liquid will not occur. However, unless a defined measure of liquid to vapor phase transition is used to make the decision, ambiguity will remain in exempting tests. Similarly, clear and defined measurable criteria should be developed as trigger points to require higher Tier level of testing in each category of toxicology testing.
B. Residues
Question: Is the science approach presented for the four major use categories reasonable for obtaining data pertinent to determining human dietary exposure? More specifically, is the decision logic for indirect food contact sanitizers reasonable and does it provide pertinent scientific data for dietary exposure testing?
The Panel finds that the scientific approach presented for the four major antimicrobial use categories (i.e. industrial processes, antifoulant coatings, wood preservatives, aquatic outdoor uses) appears reasonable for obtaining data pertinent to determining human dietary exposure. The Panel offers several comments for Agency consideration.
Specifically, the data requirements for exposure and toxicity data for sanitizers (i.e, pesticides embedded in plastic food-contact articles and those applied to food-contact surfaces, egg washes and vegetable rinses) proposed in Subpart W of 158 are similar to those that have been used by FDA for their regulation as food additives. There are certain instances, however, where the Agency is proposing more stringent requirements. Some of these requirements could provide additional data useful for assessing dietary exposure. But, they bear further consideration in light of any additional benefits to the public health, balanced with wise use of government and industry resources.
For evaluation of chronic toxicity hazards due to exposure to these substances, the Agency assumes that complete migration into food occurs over the useful lifetime of the plastic product. Estimates of the amount of food contacting the product over its service life will permit an estimate of chronic exposure without the need for costly experimental migration studies by the petitioner (However, in certain instances, migration studies may be needed). FDA has used this approach. In addition to plastics, the Agency will need to apply the same approach for sanitizers embedded in food-contact rubber articles, such as conveyor belts and gloves. EPA should revise its proposal to include references to rubber articles.
If concerns of an acute toxic hazard arise from use of a sanitizer embedded in food-contact plastic, the Agency is requiring migration studies to determine the transfer rate into the plastic. The migration studies presumably would be used to determine exposure for assessing the gravity of the concern. For a migration study, the Agency (or the petitioner or registrant) would need to design a study protocol that mimics the time/temperature food/article contact scenario. FDA has recognized, however, that significant migration is not likely to occur for food/article contact at room temperature or below for short periods of time (minutes to a few hours). A low level of migration and, therefore insignificant dietary exposure, under anticipated conditions of use should not be expected to raise concerns of acute toxicity. It would be helpful, therefore, if the Agency were to define the time frame that defines a hazard as acute and explain the basis for triggering such concerns (e.g., known neurotoxicity) for sanitizers embedded in food-contact plastics. The need for acute toxicity studies might be made a Tier I Conditional Requirement.
The Agency's new approach to regulating sanitizer formulations requires that a petitioner propose a numerical tolerance in food for the active ingredient component of the sanitizer formulation or, based on reasonable grounds, request an exemption to a tolerance. The imposition of a tolerance presents a significant burden on the Agency and the petitioner, requiring residue analysis in all foods that may contact a given sanitizer formulation and means of enforcement. Given that FDA has never required a tolerance to be established for any of the about 40 antimicrobial formulations it has regulated, it appears unlikely that the Agency will need to establish tolerances during its stewardship of sanitizer formulations. As the Agency observes, the chemicals cleared for use as sanitizers include "soaps, surfactants, chlorine or other halogen precursors, and high molecular weight polymeric materials with surfactant properties....designed to be highly water soluble .... are generally characterized by low intrinsic toxicity or toxicity which is rapidly dissipated once they come in contact with microorganisms." The Agency should consider eliminating the requirement that the petitioner address the need for a tolerance. However, the Agency should also retain its authority to require a tolerance in the unlikely event that a concern arises for the need for one.
Note (6) of Section C Table 1 of § 158.1109 states that an analytical method capable of measuring residues of sanitizer formulations in foods/feeds is required for any food use. The Table indicates that this is a Conditional Requirement, so that there may be mitigating circumstances that result in a conclusion that a method is not needed. It is well-known, nonetheless, that the development of analytical chemistry methods for analysis of compounds, particularly antimicrobials, present at extremely low concentrations in a highly complex food matrix is difficult, if not impossible, and costly. Added to this concern is the possibility that a particular formulation may be used in contact with any number of foods and food types in settings ranging from agricultural premises to public food service establishments and the proposal seems untenable. Therefore, alternative approaches to assessing the residues of sanitizer formulations transferred to food in order to evaluate safety should be considered (e.g modeling, etc.).
In this light, § 158.1108 (a) (2) (ii) of Subpart W notes that the calculation of the amount of any component of a sanitizer formulation that will transfer to food is based on Directions for Use of the formulation (i.e., the at-use concentration of the specific component) combined with "historical residue data concerning the amount of sanitizing solution remaining on food-contact surfaces." This reference to "historical" data derives from the approach that FDA has used to clear sanitizers. The 1986 FDA guidance for sanitizers (revised in 1993 with no substantive changes) stated, based on substantial experimental data, that a worst-case assumption for residual sanitizer solution on an "adequately drained" surface is 1 mg/cm2. FDA concluded that this value could be used, in the absence of residue data from the petitioner, as the basis for a conservative dietary exposure assessment by assuming all of the residual sanitizer is transferred to food. The Agency might wish to explicitly incorporate this 1 mg/cm2 into Section C Table 1 of § 158.1109, as a default surface residue for estimating sanitizer residues in food.
C. Ecological Effects/Environmental Fate
Question: The Agency believes there are eight use scenarios for which ecological risk assessments are not necessary: agricultural premises and equipment; food handling/storage establishments premises and equipment; commercial, institutional and industrial premises and equipment; residential and public access premises; medical premises and equipment; human drinking water systems; materials preservatives; and swimming pools. For these use scenarios the Agency will require only a minimal set of ecological effects and environmental fate data for use in labeling manufacturing and certain end-use products. These data are: avian acute oral LD50, acute freshwater fish LC50, acute freshwater invertebrates EC50, and hydrolysis study. Does this approach seem reasonable?
The Panel believes that the reduced data set requested for these specified uses (i.e. agricultural premises and equipment; food handling/storage establishments premises and equipment; commercial, institutional and industrial premises and equipment; residential and public access premises; medical premises and equipment; human drinking water systems; materials preservatives; and swimming pools) appears justified only if data available from other programs both within and outside the Agency are adequate to assess risk. This must be verified because as the proposed rule indicates there are some industrial uses which may result in significant, frequent releases. The use of antimicrobials may likely increase due to widespread concern over direct and indirect contamination of food, waters, and equipment in many processing activities. The assumption of minimal exposure should be verified by examining existing data sets of environmental residues of frequently monitored chemicals in addition to data from NPDES permitting/testing activities to determine the frequency of detection, concentrations, and effects of chemicals likely to be used in antimicrobial use-patterns. For some uses, these data could be used to conduct basic risk assessments based on recommended efficacious concentrations and potential loading levels based on facility size, production rates, use rates, etc.
The Panel also is concerned over the lack of chemical fate data requested. Hydrolysis will be an important fate pathway for only a subset of chemicals. Biodegradation data under both anoxic and aerobic conditions are needed to perform risk assessments. Again, the Agency indicates that these data will be available from other places both within and outside the Agency. These data sources should be provided in a tabular format which provides the test reference number used by the responsible office or agency with primacy for the data. It is recommended that data access from other Agency Offices (e.g. Office of Water) should be "seamless", or preferably, that these fate and effects data are merged with existing on-line databases within OPP. This will facilitate data access for site-specific risk assessments.
The Panel is further concerned about the lack of inclusion of any microbial testing. The proposed rule indicates that efficacy testing will be conducted for a selected group of chemicals such as sanitizers. However, additional non-target microbial data should be provided for all chemicals. These tests are needed not only to ensure the safety of environmental discharge but also for protection of POTWs and other treatment systems which often rely on microbial treatment processes. These data should be considered in addition to the basic fish and invertebrate toxicity tests.
Finally, the Panel is interested in what appropriate precautionary labeling might be used to protect fish and wildlife from improper use of antimicrobials. For traditional pesticides this may consist of a buffer zone or precautions about disposal in aquatic systems to minimize exposure. However, for many of the proposed indoor-use categories the majority of antimicrobial chemicals would enter a POTW or private sewage system through normal use patterns. It is unclear how a precautionary label would be constructed to minimize exposure to ecological resources
.
Question: The Agency believes that for the remaining four use scenarios (industrial processes and water systems; antifouling coatings; wood preservatives; and aquatic areas) we will perform ecological risk assessments (primarily, aquatic risk characterizations). Therefore, The Agency will require a tiered set of ecological effects and environmental fate data for these use scenarios (with an emphasis on water column and benthic studies addressing effects on aquatic organisms and environmental fate in these compartments). Does this decision logic seem reasonable? Further, since the data required are designed to address aquatic risks, does the Agency need to gather more information to address other risks?
The Panel finds that full ecological risk assessments are necessary for the four major-use categories (i.e. industrial processes and water systems, antifoulant coatings, wood preservatives, aquatic outdoor uses).
The Agency presentation indicated that there is concern about possible redundancy of testing requirements across Agency Program Offices. If redundancy of requirements is a concern then this should be addressed through harmonization among other Agency Program Offices of fate and effects testing where possible. Experience should indicate those cases where testing from one Agency Office may be substituted for the requirements of the Office of Pesticide Programs.
The Agency indicated that exposure data may be difficult to obtain. Exposure assessments must be generated either from modeling (e..g. based on efficacious concentrations and potential loading levels based on facility size, production rates, etc.) or actual data sets (e.g. existing data from NPDES or other monitoring activities). The number of facilities should not be a deterrent. Rather, the diversity of type and use should be examined to further categorize use patterns and exposure scenarios. This is reasonable and should be pursued for representative industries, chemicals, and uses.
Aquatic exposures should be the greatest concern. However, there are terrestrial situations where some chemicals such as wood preservatives (e.g. animal feeders, bird houses, places where high exposure could occur due to intimate skin contact, licking, chewing, etc) or anti-fouling coatings (e.g. boat yards or painting facilities where paint chipping, dust, and subsequent exposure) could present terrestrial wildlife risks. These specific cases need to be reviewed carefully by the Agency.
In terms of the conduct of ecological risk assessments which the Agency alluded to that would be primarily aquatic risk characterizations, it is unclear as to the focus or endpoints considered in this risk characterization process. This process needs to be focused and refined considering the increasing amount of data available in this area and improvements in ecological risk assessment methodologies.
A final concern, which is not unique to antimicrobials, concerns the toxicological testing of metabolites identified from degradation experiments. Full testing under the 835 Guidelines of Part 158 requires biodegradation testing under aerobic and anaerobic conditions in which primary metabolites are identified. The Agency should ensure that some mechanism exists for toxicity testing of these metabolites. Some metals and metalloids (e.g. mercury and selenium) with antimicrobial activity can be microbially methylated to more toxic by-products. Similar changes can also occur with organic materials. It may not be necessary to explicitly test every metabolite. However, it is possible to sequentially test chemicals subjected to degradation processes using designs similar to a microcosm or sediment toxicity test. Replicate series of chemical/water or sediment/water mixes could be sequentially tested over the degradation life of the chemical to determine if toxicity decreases according to anticipated loss of the parent compound. Departure from the expected decline may indicate the presence of a toxic metabolite.
D. Human Exposure
Question: Are the approaches presented reasonable for obtaining data pertinent to determining application and post-application exposure? Has the Agency adequately covered all use/exposure scenarios? For multiple exposure scenarios for one pesticide product, should the Agency require data for all exposure scenarios or for a subset of scenarios?
QUESTION 1: Are the approaches presented reasonable for obtaining data pertinent to determining application and post-application exposure?
The Agency has proposed twelve antimicrobial use categories to assist in the explication of human exposure data requirements. These categories, or general use patterns, provide the framework for subsequent data requirement tables. The Panel found that these categories provide a sensible and reasonable approach to organizing data requirements, considering the wide range of individual chemicals, possible use scenarios, and environmental and health-related endpoints. While the classification into twelve categories is complex, it should be recognized that it is a starting point for information collection; as time passes, it will probably be possible to merge and collapse categories and data elements, resulting in a simplified system. The Pest Management Regulatory agency of Health Canada has reviewed the Agency approach and found it to be generally consistent with their own approach. They note in their comments that the use of common categories will facilitate comparison of data across agencies, and opens the possibility of joint or shared reviews. During public comment the Agency indicated that a similar relationship exists with the California Environmental Protection Agency, and that the use of common categories has proven helpful in the sharing of information and in risk assessment activities. Comments submitted by the Chemical Manufacturers' Association expressed the view that the Agency categories were too broad to be useful. In public comment, the Agency indicated its awareness that these categories are broad, and that they may need to be subdivided to provide more specific guidance. The Panel believes that this greater level of detail would be appropriate for a guidance document, but not for a data requirement such as Subpart W. Therefore, the Panel endorses the use of the categories as presented by the Agency, and encourages further refinement as new information becomes available.
One additional concern is that the Agency's definition of "post-application exposures" may be too restrictive in light of the actual exposure situations. "Post-application exposures" were described as exposures to bystanders, or people who enter an area treated with pesticide. This definition would not necessarily include exposures which result from handling, or otherwise coming into contact with treated material. For example, people handling preserved wood, textiles, and leather have been demonstrated to have significant exposure to preservatives (e.g., chlorophenols).
QUESTION 2: Has the Agency adequately covered all use/exposure scenarios?
It is impossible to identify all use/exposure scenarios, but the Agency appears to have identified the most significant use/exposure scenarios.
In the discussion of Use Categories on page 6 of the Proposed Rule document, the Agency enumerates the twelve general use patterns which provide the framework for subsequent data requirement tables. The Agency then describes the use categories on pages 7-10. This description is very helpful, and should be retained. It would also be helpful for the Agency to add succinct language which would explain the Agency's rationale for each of these categories.
QUESTION 3: For multiple exposure scenarios for one pesticide product, should the Agency require data for all exposure scenarios or for a subset of scenarios?
The Agency should work initially with the full set of possible exposure scenarios for a particular product, then allow submission of data which would support combining specific scenarios, or which would allow the elimination of some scenarios where there is no documented exposure.
ADDITIONAL COMMENTS
The proposed rule states that the EPA will determine whether industrial standards for OSHA-regulated industries provide adequate protection for antimicrobial pesticides. If these standards were determined to be adequate, monitoring for uses in those industries would not be required. This provision would put EPA into the position of evaluating the adequacy of OSHA standards, which would not seem to be a wise strategy. Furthermore, even in cases where the OSHA standard is protective, it is possible (for example in a business with fewer than 10 employees) that the enforcement of the OSHA standard is so limited that the standard is not, in fact, protective. Finally, it should be kept in mind that the OSHA standards are intended to protect healthy working people who are exposed 8 hrs/day, 40 hours/week -- not the sort of long-term exposures to a range of populations which EPA must address.
The role of biomonitoring in the exposure assessment should be described more fully. While more biological monitoring has the advantage described in the proposed rule, it is most effective when used as part of a comprehensive exposure assessment strategy, which includes measurement of inhalation and dermal exposure. From an exposure prevention point of view, biological monitoring used alone has the significant limitation that it does not reveal anything about the route of exposure. This information is essential to direct preventive measures to reduce exposure. Finally, many biological markers of exposure are subject to wide inter-person variability, which makes them difficult to use and interpret in population studies.
One point which was raised but not fully explored in the discussion is the difference between children and adults when evaluating indoor residential exposures. For example, a child's exposure to antimicrobials in carpet could be very significantly different from the exposure an adult would experience in the same residential environment.
E. Efficacy
Question: Should the Agency begin using a new efficacy standard method, Hard Surface Carrier Test, when only one part of the method has been validated (i.e.,distilled water)? The remaining portions of the method (organic soil, hard water) are under collaborative study and are expected to be completed by the end of 1997.
The Panel acknowledges these are draft documents, subject to revisions as new information becomes available or better methods and procedures are developed. Concerning the use of the "Hard Surface Carrier Test" (HSCT) in full, knowing that the only component of the test to be validated to date is that of distilled water, the Panel finds that it is prudent to continue using the Use Dilution Test Method, despite its shortcomings, as the Agency awaits completion of the validation trials for the other two components of the HSCT, namely that for hard water and organic soil. These appears to be support for the development of a fully validated test with a phase-in period to allow for a smooth transition to the new method. If there is concern about drafting language into the proposed Subpart W which would allow the Agency to adopt the HSCT in the future, it could be noted that the Agency will replace the Use Dilution Test with the HSCT, pending satisfactory completion of the validation trials for hard water and organic soil.
FOR THE CHAIRPERSON:
Certified as an accurate report of findings:
Larry C. Dorsey
Designated Federal Official
FIFRA/Scientific Advisory Panel