Commenters requested clarification on the criteria and suggested the revisions would increase the burden to registrants. All of the listed flagging criteria need not apply to a toxicology study. If any of the criteria listed are applicable to the study, then the corresponding criterion number is to be included in the flagging statement submitted with the study. In the proposed rule, the Agency acknowledged that the revisions could flag more studies but this was expected because of the new types of toxicity studies to further protect infants and children. EPA made no revisions to the flagging requirements in the final rule. EPA has responded to comments in its Response to Comments document in the docket for this rule.

E. Data Waivers

EPA proposed reformatting the waiver process while retaining the provisions. Several commenters expressed their concerns about clarity, timelines and organization of information for waiver requests and made several suggestions. EPA refined data requirements and test notes to help the registrant determine if a waiver request is in order. Applicants are encouraged to discuss the waiver with the Agency before developing and submitting supporting data, information, or other materials. The Agency is committed to timely decisions and notification of the applicant. Organizational changes that were proposed will be retained for the final rule. EPA has responded to comments in its Response to Comments document in the docket for this rule.

F. Formulators' Exemption

EPA proposed to remove or revise provisions in part 158 that directly or indirectly arise from the statutory formulators' exemption of FIFRA sec. 3(c)(2)(D). First, EPA proposed to remove language in Sec. 158.50 pertaining to the statutory formulators' exemption. Second, EPA proposed removing the asterisks denoting the application of the formulators' exemption to product chemistry and toxicology data requirements.

A number of commenters objected to the removal of formulators' exemption language, and others were confused by the removal of the asterisks. It is clear that commenters are confused by the distinction between the array of data that the Agency must have to determine whether a pesticide may be registered (the data requirements of part 158), and the means by which those data requirements are satisfied (the data citation and compensation provisions of part 152, subpart E, including the
[[Page 60937]]
formulators' exemption). In short, part 158 specifies the ``what'' and part 152 specifies the ``how'' of data requirements.

The primary purpose of part 158 is to specify the data requirements pertaining to a pesticide product. Part 158 was never intended to serve the broader purpose of specifying the various means by which an individual applicant can legally satisfy the data requirements: that is the purpose of the data compensation provisions of part 152. Part 152 explains all of the means of satisfying a data requirement specified in part 158, including submitting new data, citing existing data, citing to public literature, obtaining a waiver, or claiming eligibility for the formulators' exemption. EPA believes that it should reinforce this distinction by removing from part 158 what is actually incomplete information about the formulators' exemption.

Eligibility for the formulators' exemption is not a function of a data requirement. Rather, eligibility depends on the purchase of a registered product for incorporation into another product. The 1984 regulations erred in attempting to apply the formulators' exemption to specific product chemistry and acute toxicology requirements by means of the asterisk notation. First, the manner in which the asterisks were displayed was such that it was not clear precisely when the formulators' exemption did and did not excuse an applicant from the requirement to submit data. Further, it was unclear because it potentially conveyed the notion that the data requirement need not be satisfied. The fact that certain data need not be submitted or cited by an applicant eligible for the formulators' exemption does not mean that those data are not necessary to support the registration of the product, merely that the data requirement has been satisfied by another means. Usually the requirement has been satisfied by submission of data by the producer of the registered TGAI or manufacturing use product (MP) that the applicant purchases.

Additionally, maintaining information on the formulators' exemption in two locations in the Code of Federal Regulations is administratively cumbersome. As one commenter noted, the statute has been revised since both of these regulations were issued, and neither Sec. 152.85 nor Sec. 158.50 is accurate or complete. For this reason, EPA believes it is important to consolidate the formulators' exemption language in a single location.

All commenters correctly pointed out that although EPA indicated in the preamble that the formulators' exemption text of Sec. 158.50 was to be relocated to part 152, no proposed regulatory language was included. EPA agrees that it did not include in the proposal the actual regulatory text that would be incorporated into part 152. In a companion final rule making technical changes, and which is published elsewhere in this issue of the Federal Register, EPA has included the revised language, which would incorporate the provisions of Sec. 158.50 into Sec. 152.85 with needed conforming text changes. EPA has also corrected Sec. 152.85 to reflect current FIFRA sec. 3(c)(2)(D), as amended in 1988. Except where required as a result of these statutory amendments, EPA has made no substantive change to the exemption or EPA's interpretation of its applicability.

Although EPA believes that the formulators'exemption should properly be located in part 152 together with other provisions concerning submission or citation of data, the Agency recognizes the value of referring to the provisions of part 152 in part 158. Accordingly, EPA has revised Sec. 158.70(a), by including a new paragraph (1) which explains that the provisions of part 158 should be read in conjunction with those of part 152, subpart E.

G. Minor Uses

EPA proposed to delete material in Sec. 158.60 concerning minor uses. Minor use policies in existence in 1984 and information in anticipation of reregistration needs for data were included in original part 158. The information is by no means complete concerning EPA policies on minor uses, which have since expanded by statute. Nonetheless, several commenters wanted EPA to retain the information in paragraphs (a)(2) and (3). EPA has in the final rule retained paragraphs (a)(2) and (3), but has removed the remaining material and renumbered those paragraphs. The paragraphs being deleted have been superseded (the definition in paragraph (a)), are guidance only (paragraphs (a)(1) and (b)), or are covered by regulations elsewhere (paragraph (a)(4)).

H. WeightofEvidence Approach

The weightofevidence approach is referenced in part 158 under several disciplines. The approach requires a critical analysis of the entire body of available data for consistency and biological plausibility. Some considerations in this approach are listed below:

In the case of the developmental neurotoxicity (DNT) study, such a weight of the evidence approach is used when evaluating:

1. Treatmentrelated neurological effects in adult animal studies, such as:

2. Treatmentrelated neurological effects in developing animals, following pre and/or postnatal exposure, such as:

3. Causative association between exposures and adverse neurological effects in human epidemiological studies

4. A mechanism that is associated with adverse effects on the development of the nervous system, such as:

A compound could be subject to a DNT requirement under a variety of circumstances using these criteria in a weight of evidence approach that considers dose response, logical pattern of effects, data quality, biological plausibility, consistency of observations in the broader toxicological database, likeness of the case to structural analogues, and mode of action understanding. For example, the following scenarios for 3 different chemicals (chemicals A, B, and C) describe findings that could lead to the conclusion that a DNT study is needed. Chemical A is found to result in
[[Page 60938]]
responses consistent with an effect on the central nervous system (CNS): staggering (i.e., abnormal gait) at the mid and high doses and convulsions at the high dose are seen in a study, and abnormal gait at the mid and high doses and cortical lesions in the brain at the high dose are seen in another study. Chemical B is a GABA (gamma aminobutyric acid) receptor antagonist (i.e., a CNS mode of action that block inhibitory systems that are involved in nerve responses) and is found to result in functional effects in the animal studies consistent with this mode of action, such as hyperactivity, altered response to sudden loud noises, and seizures (only at very high doses). In developmental toxicity studies, Chemical C results in doserelated microcephaly, a rare finding indicative of the brain neurons not proliferating normally.

However, a single effect would not necessarily always trigger a DNT. For example, a small decrease in brain weight at the highest dose tested in one adult animal study but no indications of neurotoxicity, including the lack of corresponding decreases in brain weight in other adequate toxicity studies, would not necessarily trigger a DNT. Similarly, a decreased response to stimuli at doses that result in significant body weight loss and poor health of the animal may not provide a weightofevidence basis for triggering the DNT.
VI. Discussion of Key Comments on the Data Tables (Subpart B) A. Use Patterns

EPA proposed subdividing the current nine major use patterns to 15 major use patterns to fully address nonagricultural uses. Commenters asked for definitions of the proposed major use patterns and the phrases ``major use pattern,'' and ``pesticide use site groups.'' One commenter suggested adding a new major use pattern in addition to the ones proposed by EPA. Commenters also identified inconsistencies in major use patterns between the preamble and the regulatory text. EPA believed that the resulting use patterns from the subdivision of existing major use patterns were fairly selfexplanatory and believed that adding the suggested terrestrial nonfood noncrop uses might create too fine a distinction and add to the already existing confusion. However, the Agency does appreciate the commenters' assistance in locating inconsistencies between the regulatory text and the preamble and believes the inconsistencies have been corrected.

One major use pattern in the proposed rule, Indoor medical, has been eliminated from the final rule. It is a use pattern primarily applied to antimicrobial products, not conventional pesticides, and will be considered for subpart W when proposed for comment. There were several variations of aquatic nonfood use patterns that commenters found confusing. The definition of the aquatic nonfood residential category was questioned by several commenters who assumed it referred to indoor tropical fish aquaria or koi fish ponds in yards. A survey of labels associated with this use category produced only a handful of products. Therefore EPA has consolidated the various aquatic nonfood use patterns into one aquatic nonfood use pattern, thus reducing the number of aquatic nonfood patterns to one. The elimination of Indoor medical and several aquatic nonfood use patterns reduced the final number of major use patterns. Thus, the final number of major use pattern for conventional pesticides will be 12, rather than the 15 in the proposed rule. The final 12 use patterns are: terrestrial food crop; terrestrial feed crop; terrestrial nonfood crop; aquatic food; aquatic nonfood; greenhouse food crop; greenhouse nonfood crop; forestry; residential outdoor; residential indoor; indoor food; and indoor nonfood.

In addition, not all the general use patterns will appear in the data table for each discipline. Some of the use patterns have been collapsed under a larger major use pattern for ease of use. For example, the major use patterns in the Toxicology Data Requirements table consist of Food and Nonfood. The discussion in Sec. 158.500(b) explains that the general use patterns of terrestrial food crop, terrestrial feed crop, aquatic food, greenhouse food crop, and indoor food have been placed under the major use pattern Food. The Nonfood use patterns include products classified under terrestrial nonfood crop, aquatic nonfood, greenhouse nonfood crop, forestry, residential outdoor and indoor, and indoor nonfood. Therefore only two major use patterns appear in the data requirement table for Toxicology. Similar adjustments have been made to other disciplines as appropriate. B. Appendix A

EPA proposed updating the current Appendix A, a compendium of pesticide use sites associated with major use patterns to assist registrants in determining which data requirements might apply to their products. EPA also proposed removing the updated Appendix A from 40 CFR part 158 and placing it on the OPP website and titled as Pesticide Use Site Index. This change in location would allow EPA to correct and update the pesticide use sites with some regularity without a complicated and lengthy rulemaking. Commenters either wanted to retain Appendix A in 40 CFR part 158 or were in favor of posting it on the OPP website. The latter were more concerned that the information be updated and revised more frequently. Since Appendix A is meant to be an index of pesticide use sites and major use patterns but not a requirement for applicants, EPA believes that it is more properly posted on the OPP website to assist applicants in locating the relevant pesticide use site(s) and the corresponding data requirements. Users are encouraged to submit comments and suggestions to the contacts listed on the Web page. OPP will update the Pesticide Use Site Index on a timely basis to keep the information current for users. Accordingly in the final rule, EPA has removed Appendix A from 40 CFR part 158. The information in the current Appendix A has been updated, titled Pesticide Use Site Index, and is available at http://www.epa.gov/pesticides/regulating/registering/ datasources.htm.

C. Test Substances

EPA is continuing its longstanding system of identifying test substances in the tables as follows: Technical grade of the active ingredient (TGAI); manufacturinguse product (MP); pure active ingredient (PAI); pure active ingredient, radiolabeled (PAIRA); enduse product (EP); and typical enduse product (TEP).

D. Required and Conditionally Required Data

Some commenters were confused by the explanations of R and CR in the proposed rule and requested tighter definitions and clarification of the test notes since the latter provided insufficient guidance. In the proposed rule, EPA requested comment on its R/CR designation, and received no suggestions for alternative means of presenting the data requirements. As described in the preamble to the proposed rule, the R/ CR terminology is a general presentation of the likelihood that a data requirement will apply. The use of R does not necessarily indicate that a study is always required, but that it is more likely to be required than not. The use of CR means a study is less likely to be required. However, both R and CR designations must be read in the context of the accompanying test notes to provide context for the R/CR in the table. An applicant may assume that a data requirement with R will typically [[Page 60939]]
be required all the time. The test notes accompanying that R designation may provide supplementary information or identify some condition(s) when the study is not required. A CR designation will generally include more extensive test notes describing the limited conditionality of the requirement. The final rule continues this longstanding practice. EPA revised some of the test notes to clarify the conditions under which the data would be required.
VII. Discussion of Key Comments on Identifying Data for Experimental Use Permits (EUPs) (Subpart C)

EPA requested comment on a way to identify data requirements for EUPs to replace the current bracketing system within each data table. A commenter suggested that EPA should separate out the data requirements applicable to experimental use permits, which have been expressed since 1984 by simply bracketing a registration data requirement in the tables. Other commenters misunderstood the bracketing, assuming that bracketed data requirements were somehow conditional in nature. EPA agrees that the bracket system diminishes the visibility of the EUP data requirements and leaves them scattered throughout the registration data requirements, and has therefore separated out and consolidated them. At the same time, EPA has updated the test notes to reflect those in the subparts on registration data requirements.

Because an experimental use permit is intended to precede a full registration, EPA has elected to place those data requirements early in the part 158 organizational structure. An alternative location for EUP data requirements would have been to locate them in part 172, thereby consolidating all EUP requirements in one place. However, examination of part 172 yielded no logical location for the data requirements except at the very end. Accordingly EPA has placed EUP requirements in subpart C of part 158, preferring to keep all data requirements pertaining to conventional pesticides in one place for ease of use. Where test notes for registration requirements have been revised based on comments to the proposed rule, in separating out EUP requirements, EPA has also revised those same test notes as they apply to EUPs. VIII. Discussion of Key Comments on Product Chemistry Data Requirements (Subpart D)

EPA proposed a few changes in product chemistry requirements and it received a number of comments on elements of the data requirements that EPA had not proposed changing. They include:

These comments are outside the scope of the proposal and may be considered for future revisions of part 158. Accordingly, EPA has not revised the final rule.
IX. Discussion of Product Performance Data Requirements (Subpart E)

EPA has transferred the contents of the product performance section (current Sec. 158.640) essentially unchanged into the revised part 158. The regulatory text of the product performance section is reprinted in this final rule for clarity and completeness.
X. Discussion of Key Comments on Toxicology Data Requirements (Subpart F)

A. Data Requirements

1. Immunotoxicity. EPA proposed requiring functional immunotoxicity testing to evaluate the potential of a chemical to adversely affect the immune system since immune system suppression has been associated with increased incidences of infections and neoplasia. While the Agency understands that traditional subchronic and chronic rodent studies can provide much useful information on certain immunological endpoints such as hematology, lymphoid organ weights and histopathology, these studies do not provide a full and integrated evaluation of immune function. As a result of recommendations from the National Research Council (NRC) review and the FIFRA Scientific Advisory Panel (SAP), the Agency proposed requiring functional immunotoxicity testing along with the data from endpoints in other studies to assess the potential risk of pesticides on the immune system more fully.

Fifteen commenters submitted a variety of comments on this data requirement. All comments are addressed in the detailed Response to Comments document in the docket. Key comments are discussed in this unit.

Two commenters requested clarification of when this testing would be required and one commenter compared the U.S. requirement with that of the European Union (EU). Three commenters strongly supported including immunotoxicity testing in the toxicology data requirements for all pesticides. Six commenters opposed the codification of this data requirement on several bases and offered alternatives: divergence in immunological structure and response between species that gives animal studies limited predictive power for immunogenicity in humans; using data from other toxicity studies as a trigger for immunotoxicity studies; and changing from R to CR. EPA disagrees with these comments because data and analysis have shown that functional immunotoxicity testing, particularly when considered in conjunction with data already required by EPA on immunotoxic endpoints, is likely to increase EPA's ability to identify pesticides with immunotoxic effects. Additionally, functional immunotoxicity testing allows for better characterization of the possible effects of an immunotoxicant.

Three commenters had detailed technical questions about the test guideline which were not appropriate for discussion in part 158 since the latter concerns only data requirements. Their comments and suggestions were forwarded to the appropriate scientists for review and consideration in the context of guideline revision. While EPA agrees that the testing protocol may need further refinement, discussions on alternative testing paradigms will continue through the various scientific venues (e.g., International Life Sciences Institute/Health and Environmental Sciences Institute (ILSI/HESI) cooperative effort) as well as through future consultation with stakeholders on the development and validation of this test guideline.

EPA recognizes that there are a range of opinions on the necessity of an acrosstheboard requirement for functional immunotoxicity testing. However, EPA's judgment, as supported by the recommendations of the NRC and FIFRA SAP, is that there is valueadded from requiring functional immunotoxicity testing for all pesticides. Therefore in the final rule, EPA retains a requirement for immunotoxicity testing on all food and nonfood pesticides on the TGAI. EPA has responded to comments in its Response to Comments document in the docket for this rule.

2. Prenatal developmental toxicity. EPA proposed amending the name of the requirement to correspond with the current terminology and to require two species for all nonfood pesticides. Commenters suggested making this requirement conditional based on results of other Tier 1 studies or on a
[[Page 60940]]
likely exposure pattern. EPA proposed requiring a second species because it believes the data will provide some assurance that the Agency will not be basing an assessment on a single species that might be highly sensitive (or the opposite) when compared to another. The final rule will maintain these changes to adequately characterize potential hazards to pregnant women and their fetuses.

3. 21day dermal and 90day dermal toxicity. EPA proposed a 21 to 28day dermal toxicity test for all food use pesticides since it is generally needed for worker risk assessments. Analyses of exposure information have shown that this duration of exposure is typical for agricultural workers in various components of their job. EPA proposed not requiring the 21 to 28day dermal toxicity test for nonfood uses. However, if the dermal route is the primary route of exposure for nonfood uses, a 90day study would be required because EPA believes the 21 to 28day subchronic dermal toxicity test is insufficient to identify potential hazards.

Several commenters questioned requiring a 90day study for nonfood uses when exposures rarely exceed 45 days. EPA considers the 21 to 28 day dermal study insufficient for nonfood use assessment because higher tiered oral studies (i.e., chronic or carcinogenicity studies) are not usually required for nonfood use pesticides. While 45day exposures are common, EPA believes that they are not the maximum duration. For example, professional applicators may be subjected to repeated exposures during the 3 months of peak summer infestations. Since for many pesticides there is increased toxicity with increased exposure, professional applicators may not be adequately protected with 45day studies. Existing regulations provide flexibility to implement alternative studies, on a casebycase basis, as appropriate. Registrants should consult with the Agency if there is any question regarding the appropriate duration of the study. The highest level of hazard evaluation available for a nonfood use pesticide is satisfied through a subchronic toxicity test, i.e., a 90day repeated exposure to the nonfood pesticide. Therefore, the final rule will require the 90 day dermal toxicity study for nonfood uses.

4. Reproduction and fertility effects. EPA proposed to require a reproduction study for nonfood uses but emphasized that the requirement is based on potential exposure. Commenters requested further clarification when the study would be required. Requiring the study for nonfood use pesticides would be based on a weightofevidence consideration of the toxicology data and potential exposure in terms of the frequency, magnitude, and/or duration. This is primarily an exposurebased data requirement and will not always be necessary. Registrants should consult with the Agency if there is any question whether the study must be conducted.

5. Developmental neurotoxicity (DNT). EPA proposed that developmental neurotoxicity testing (DNT) be conditionally required for food and nonfood use pesticides. Thirteen commenters were unclear about the conditionality of this requirement and requested clarification about Test Note 27. Test Note 27 identified the effects to be considered in the weightofevidence approach.

One commenter questioned whether the results of standard tests in developing animals were sufficient to trigger a DNT test and whether the inhibition of cholinesterase activity (ChEI) would be the most sensitive effect for organophosphorus and Nmethyl carbamate pesticides. The Agency has completed review of 20 DNT studies conducted with organophosphorus pesticides. In 13 out of 20 studies, ChEI was measured in the pups; cholinesterase was the most sensitive endpoint in those 13. Only a limited number of DNT studies are available for carbamates, and the endpoint for only one chemical was used to assess acute dietary risk.

Two commenters suggested amending the 2generation reproduction study to include findings of thyroid effects, thus providing another criterion for DNT testing. Although such a criterion was included in the proposed weightofevidence approach, experience gained with the study resulted in the removal of this criterion. Instead, when thyroid effects of concern are observed, the Agency may require a more specific special study. In the final rule, EPA continues to encourage registrants to conduct DNT studies in combination with a 2generation reproduction study when addressing the DNT requirement.

Ten commenters asked for clarification of Test Note 27 to indicate whether the listed effects were part of the approach and not individual triggers. EPA has revised this Test Note to eliminate the impression that the items in the list were individual triggers and referred commenters to its published Risk Assessment Guidelines for a more detailed explanation of the terms used in the test note. Due to an addition of a test note, Test Note 27 in the proposed rule was re numbered to Test Note 28 in the final rule.

Therefore, the Agency is conditionally requiring the DNT study in the rat for food and nonfood pesticides. All available toxicology data for the pesticide will contribute to the weightofevidence
determination of the need for a DNT study. The criteria for the weight ofevidence determination are listed in Test Note 28 and include neurological effects from adult animal studies as well as
neurobehavioral effects after pre and postnatal exposure of the pesticide to young animals.

6. Scheduledcontrolled operant behavior, peripheral nerve function, and neurophysiology sensory evoked potentials. Commenters wondered if these tests would be commonly required and requested specific triggers for these studies. EPA discovered upon review that these studies were seldom required during the reregistration process and determined the studies could be removed from the table of commonly required studies. If the need arises in the future, the Agency may require any of these studies on a casebycase basis. Validated OPPTS guidelines are in place.

7. Nonrodent chronic studies (1year dog study). In the proposed rule, EPA considered eliminating the requirement because evidence from the published literature was consistent with EPA's belief from its reviews that the study may not be needed. EPA currently requires a 90 day dog study and a 1year dog study for all food and nonfood uses to fulfill the nonrodent data requirements. EPA referenced published literature that suggested that the 1year dog study may not be necessary. Based on a retrospective analysis of a large body of 1year dog studies in its toxicology database, EPA proposed to eliminate the 1year dog study but retain the 90day study. EPA solicited review and comment by the FIFRA Scientific Advisory Panel (SAP) on the results of the preliminary analysis for reference dose (RfD) derivation on May 5 6, 2005 [Ref. 10].

The FIFRA SAP reviewed the Agency's retrospective analysis of the toxicity studies and encouraged the Agency to continue its analysis with a larger database. The FIFRA SAP made the following

recommendations:

i. Increase the robustness of data analysis by including dog study datasets that were not used for the RfD determination.

ii. Conduct an analysis more representative of a prospective comparison through delineating the 13week No Observed Adverse Effect Levels (NOAELs) and Lowest Observed Adverse Effect Levels (LOAELs) [[Page 60941]]
independent of the 1year study and establish data review criteria.

iii. Consider data analysis for separate classes of pesticides.

iv. Include additional background information on RfD that provides better perspectives for reviewing the Agency position paper.

v. Revise the title of the Agency position paper to reflect the purpose of the data analysis.

The FIFRA SAP said in its report that ``if the results of the analysis continue to indicate little added value from the 1year dog studies, the Agency could move toward eliminating them on a stronger basis.''

In response, EPA conducted a more extensive analysis of dog toxicity studies on 110 chemicals representing over 50 different classes of pesticides [Ref. 12]. EPA concluded from this analysis that extending a dog toxicity study beyond a 13week duration does not provide additional essential toxicity information; eliminating the 1 year dog toxicity study does not compromise the data needed for the determination of chronic RfDs and margins of exposure (MOE). Thus, reliance on the required chronic rodent studies, 2generation rat reproductive study, and the 13week dog toxicity study provides an adequate basis for chronic RfD derivation in pesticide risk assessment.

EPA acknowledges that there may be situations where a longer duration dog toxicity study may be warranted when a pesticide chemical is highly bioaccumulating (e.g. builds up in body fat) and is eliminated so slowly that it does not achieve steady state or sufficient tissue concentrations to elicit an effect during a 90day study. EPA anticipates that this situation will be infrequent since current pesticides are not usually designed to be highly persistent and bioaccumulating. If such a chemical is encountered, EPA would require the appropriate Tier II metabolism and pharmacokinetic studies to more precisely evaluate bioavailability, half life, and steady state to determine if a longer duration dog toxicity study is needed. The circumstances that might lead to a request for the 1year dog study are identified in Test Note 36.

B. Alternative Testing Paradigms

In the proposed rule published March 2005, EPA discussed the work underway on alternative testing paradigms by the International Life Sciences Institute (ILSI)/Health and Environmental Sciences Institute (HESI). EPA is in conceptual agreement with the ILSI/HESI philosophy of moving toxicology testing away from a rigid guidelinebased screening approach and towards a more knowledgebased approach. The ILSI/HESI approach was published in a series of papers in the January 2006 issue of Critical Reviews in Toxicology.

Eleven commenters addressed the ILSI/HESI testing paradigm, all supporting its development and early adoption. One commenter suggested that EPA update the proposed rule with the ILSI/HESI study findings and reissue a revised proposed rule for comment. In a similar vein, another suggested incorporating a timetable into the final rule for modifying subpart F (Toxicology). Another commenter believed a number of the concepts developed in by ILSI/HESI were ripe for incorporation into pesticide testing requirements at this time. This same commenter suggested not finalizing the proposed rule until there was an opportunity to consider and incorporate the important concepts developed by Agricultural Chemical Safety Assessment (ACSA). EPA believes that incorporating the concepts into the final rule is premature since EPA has not had the opportunity to determine if the new testing paradigm will meet its risk assessment needs. EPA believes that delaying the remaining proposed changes which comprise the bulk of the proposal would be a disservice to the regulated community. In a differing view, a commenter was concerned about the lack of public interest representatives in ILSIEPA discussions and recommended that EPA terminate its collaborative working relationship with ILSI and industry trade groups. Since the Agency is interested in more efficient risk assessment paradigms, it will continue to work with all stakeholders in investigating efforts in that direction and welcomes the participation of any public interest representatives in the discussions.

EPA is committed to moving towards a more efficient and refined testing/risk assessment paradigm. Given the Agency's experience with regulating pesticides over the last 30 years, the Agency is interested in improving certain aspects of the testing process. In particular, EPA is more attuned to risk assessment needs (i.e., an integrated approach) that avoids requesting data not used in risk assessment and that reduces and refines the use of laboratory animals.

In the proposed rule, EPA discussed the relevance and importance of the ILSI/HESI project, Agricultural Chemical Safety Assessment (ACSA): a Tiered Approach. This project, with the participation of EPA scientists, represents a pursuit of a more efficient and accurate tiered testing of pesticide chemicals. A series of reports authored by ILSI/HESI was published in a special edition of the Journal of Critical Reviews in Toxicology in January 2006, Volume 36, Issue 1 [Refs. 1, 2, 3 and 5], summarizing their findings and initial recommendations.

ACSA represents the first comprehensive effort to scientifically redesign the toxicology animaltesting framework for agricultural chemicals. The ACSA proposal is consistent with EPA's direction and goals to develop a more efficient and reliable testing paradigm. Under the ACSA scheme, some studies would be eliminated while endpoint coverage would be increased in redesigned studies based on responses observed in a core set of toxicity tests. The value of the scheme is that animals are more fully utilized and the need for some tests can be eliminated if the core set of tests or existing knowledge does not indicate a concern. Decisions on next steps must be made throughout the course of the study as a thorough evaluation of all available information, including data on the pharmacokinetics and mode of action of the pesticide (if such data exist), could lead to different conclusions regarding the appropriate way to approach testing.

For example, in the case of the developmental neurotoxicity study, for some chemicals, it might be concluded that adequate testing of the developing nervous system would be best accomplished with a standard developmental neurotoxicity study. Refinements to the guideline study could include, for example, changes to the route and/or duration of exposure (e.g., initiation of dosing to maternal animals prior to gestation day 6, or direct gavage administration to pups during lactation), the evaluation of appropriate biomarkers of exposure or effect, the use of more targeted functional, behavioral, or cognitive testing in offspring, or the histopathological and/or morphometric evaluation of particular regions of the central or peripheral nervous system that are known to be affected by either the chemical or chemical class. For other chemicals, the information in the toxicological database could lead to the conclusion that an alternative test should be performed instead of a guideline developmental neurotoxicity study. Alternative chemicalspecific methods could be identified as a preferred option.

EPA has multiple activities underway to address the remaining science and policy issues associated with the ACSA proposal. One essential step towards
[[Page 60942]]
adopting the ACSA proposal will be conducting retrospective and prospective data analyses to determine whether this new testing paradigm will meet EPA's risk assessment needs as defined by statute. To this end, the Office of Pesticide Programs is currently working with EPA's National Center for Computational Toxicology (NCCT) to populate a Toxicological Reference Database (ToxRef). The current priority is to populate ToxRef with data from the rat 2generation reproductive study, prenatal toxicity, and systemic toxicity studies on hundreds of pesticides that represent different classes, modes of action, and toxicity profiles. EPA will use this relational database to determine the value of endpoints currently evaluated in risk assessment (i.e., the F1 versus F2 responses). This analysis will provide scientific support for EPA's adoption of the proposal as the analysis will subject the ACSA proposal to a much broader set of chemicals than that used to develop the proposal.

Another critical step is gaining scientific consensus on the triggers (i.e., the points at which a concern is indicated and a higher level of testing is needed). The retrospective analyses will also be used to refine or confirm the ACSA proposed triggers for test decisions. Once the analysis is complete, EPA will be able to complete draft guidance on testing. The analyses and guidance are planned to be subject to SAP review and public comment in 2008.

Another essential step is testing how the ACSA scheme works in practice. There are plans to conduct several case studies using the ACSA tiered testing proposal. From these case studies, EPA will be able to assess the laboratory testing feasibility of such a complex study and to evaluate the ability of the approach and its parameters to characterize known toxicants and address risk assessment needs. Based on early scientific reviews, EPA scientists are already working on improvement of the ACSA tiered testing approach.

EPA will consider the results of the SAP review of the retrospective analyses and draft guidance, issues raised by stakeholders, and the case studies, in determining what revisions to current data requirements and testing guidelines may be appropriate. As the science issues are adequately vetted and crucial questions resolved, EPA will promulgate the appropriate regulatory changes on a timely basis. In the meantime, the existing regulations provide flexibility to implement any updated, new or novel testing schemes, on a casebycase basis, as appropriate, until the changes are codified. Casebycase determinations would be made in consultations with the Agency without the necessity of the waiver process.

It should be noted that ACSA is only one proposal that EPA will consider in improving the risk assessment process of environmental chemicals. Other relevant activities to consider include the National Academy of Sciences (NAS) recommendations on Toxicity Testing and Assessment of Environmental Agents expected in 2007 (Project ID BESTU 0308A at http://www8.nationalacademies.org/cp/projectview.aspx?key=74 ), Organization for Economic CoOperation and
Development (OECD) Integrated Approaches to Testing and Assessment (http://www.oecd.org/document/42/0,2340,en264934377362835621 111,000.html), as well as predictive toxicity tools (QSAR, omics, etc.) being developed by EPA's Office of Research and Development (ORD) Computational Toxicology Program (http://www.epa.gov/comptox). With regard to the OECD effort, EPA is currently playing a leadership role in planning a workshop scheduled for December 2007. The workshop will evaluate the current state of science and regulatory programs to evaluate pesticide inert ingredients and active ingredients using the data derived from in silico (performed on computer or via computer simulation), in vitro, and shortterm in vivo models and bioassay systems.

Before considering regulatory changes to reflect the results of EPA's consideration of ACSA, NAS, and other recommendations, the Agency will develop scientific position papers on the new approach and recommendations for internal and external review. Internal review includes review by the FIFRA SAP and opportunities for public comment. External peer review as well as acceptability by other national and international regulatory authorities are crucial before implementation of any new testing paradigm and data requirements. Harmonization with the data requirements of these same authorities is also an important factor. International regulations currently require studies that were omitted in ACSA; this would pose significant problems for registrants if a harmonized approach is not adopted worldwide.

Lastly, EPA is committed to review part 158 data requirements frequently to incorporate new science that has been fully documented and peer reviewed.
XI. Discussion of Key Comments on Ecological Effects Data Requirements (Subpart G)

A. Generic Issues

EPA received comments in several areas that were common to all science disciplines under this subpart.
1. Data harmonization and lack of availability of current guidelines. The Agency received several comments stating that the data requirements for nontarget terrestrial and aquatic organisms, plants and environmental fate testing should not be promulgated if the test guidlines upon which the data requirements rely are not finalized. The Agency recognizes the importance of the connection between these data requirements and the guidance documents that provide information on how the data requirements may be satisfied. The Agency is in the process of updating its nontarget plant test guidelines with the OPPTS and the OECD. The terrestrial and aquatic animal guidelines are scheduled to be finished and available to the public by late 2007. Nonetheless, guidelines are guidance documents only, and the promulgation of data requirements does not depend on the availability of guidance documents.

2. Elimination of species names in the test notes. EPA eliminated the inclusion of preferred species names from the data requirements in subpart G. This does not represent an actual change in the requirements. Rather, the Agency determined that the indication of preferred species is a matter of guidance and should not be part of the requirements document. Species names are covered in the Agency's test guidelines, which are cited in the data requirements tables.

3. Independent laboratory validation. Concerns were raised by some commenters that the requirement to now have independent laboratory validation (ILV) of the chemistry methods used for residue measurements in the ecological and environmental fate field studies would add cost and time to these studies. They view these studies as already required and conducted under Good Laboratory Practice Standards (GLP) in 40 CFR part 160 for other data requirements. However, GLP Standards do not require an ILV. The requirement for an ILV has been in effect since the 1990s and, as such, is a codification of current practice. The ILV, as well as the original method validation, should be conducted under the GLP.
[[Page 60943]]

B. Data Requirements

1. Terrestrial organisms. i. Acute oral toxicity test with a passerine species. EPA proposed to require a second avian acute oral study on a passerine species (i.e., redwinged blackbird) to support all outdoor uses, including residential outdoor uses. The other avian acute oral study must be conducted on either a waterfowl or an upland gamebird, which has been standard policy. This revision of the avian acute toxicity data requirement elicited a significant number of comments. The comments not only concerned the addition of a passerine species to the avian acute oral data requirement, but also the test note which specifically named the redwinged blackbird (Agelaius phoeniceus) as the preferred passerine species. Some commenters suggested that the passerine requirement should be based on the results of either the mallard or bobwhite acute oral test results. They based their concerns on the fact that the redwinged blackbird is a wild species, and is not reared in a laboratory, unlike some commonly tested passerines as the canary and zebra finch. Because it is a wild species, the laboratories must request permits from the U.S. Fish and Wildlife Service (USFWS) to trap the birds. Al

FOR FURTHER INFORMATION CONTACT For information on the data requirements for ecological effects and environmental fate, contact: Ann Stavola, Field and External Affairs Division (FEAD), Office of Pesticide Programs (OPP) (7506P), Environmental Protection Agency, 1200 Pennsylvania Avenue NW, Washington, DC 20460; telephone number: (703) 3055354; fax number: (703) 3055884; email address:
stavola.ann@epa.gov . For all other questions, contact: Vera Au, FEAD (7506P), OPP, Environmental Protection Agency, 1200 Pennsylvania Ave., NW., Washington, DC 204600001; telephone number:(703) 3089069; fax number: (703) 3055884; email address: au.vera@epa.gov.