July 27, 1989

This Preamble was converted into electronic format by Richard Reich, A.T. of the Saskatchewan Research Council, 15 Innovation Blvd., Saskatoon, Saskatchewan, Canada S7N 2X8 (Reich@src.sk.ca). The document was converted into html and indexed by Michael E. Mispagel, Ph.D., College of Veterinary Medicine, The University of Georgia, Athens, GA 30602-5706 (Mispagel.M@calc.vet.uga.edu).

Environmental Protection Agency

SUMMARY: EPA is issuing this final rule that expands the regulations to require compliance with Good Laboratory Practice (GLP) standards for testing conducted in the field and for such disciplines of testing as ecological effects, chemical fate, residue chemistry, and, as required to be submitted by 40 CFR 158.640, product performance (efficacy testing). EPA is amending these regulations to ensure the quality and integrity of all data submitted to EPA in conjunction with pesticide product registration, or other marketing and research permits. EPA is also amending the FIFRA GLP standards to incorporate many of the changes made by the Food and Drug Administration (FDA) to its GLP regulations (52 FR 33768, September 4, 1987; 21 CFR Part 58).

DATE: Effective: This rule becomes effective on October 16, 1989.

Compliance: All studies conducted, initiated, or supported after the effective date of this rule shall be subject to these regulations.

FOR FURTHER INFORMATION CONTACT:

Stephen Howie, Office of Compliance Monitoring (EN-342), Rm. E-707B, 401 M St., SW., Washington, DC 20460, Telephone:(202) 382-7825.

SUPPLEMENTARY INFORMATION:

Following is an Index to the remainder of this preamble with "Back to Index" buttons at the end of each section.

Or, go back to UGA's QAU Home Page

I. Introduction

A. Legal Authority.
B. Background.
C. Consistency With FDA GLP Regulations.
D. Publication of the Complete Rule.

II. Summary of Comments and Responses

A. General Provisions.

1. Scope
2. Definitions

B. Organization and Personnel.

1. Testing facility management
2. Study director
3. Quality assurance unit

C. Facilities.

1. General
2. Test system care facilities
3. Test system supply facilities
4. Facilities for handling test, control, and reference substances

D. Equipment.

1. Maintenance and calibration of equipment

E. Testing Facilities Operation.

1. Standard operating procedures
2. Animal and other test system care

F. Test and Control Substances.

1. Test, control, and reference substance characterization
2. Test, control, and reference substance handling
3. Mixtures of substances with carriers

G. Protocol For and Conduct of A Study.

1. Protocol - general
2. Physical and chemical characterization studies

H. Records and Reports.

1. Reporting of study results
2. Storage and retrieval of records and data
3. Retention of records

III. Regulatory Requirements

A. Executive Order 12291.
B. Regulatory Flexibility Act.
C. Paperwork Reduction Act.

I. INTRODUCTION

EPA is amending the FIFRA GLP standards (40 CFR Part 160) to incorporate many of the changes made by the Food and Drug Administration to its GLP regulations.

A. Legal Authority

These standards are promulgated under the authority of section 3, 4, 5, 6, 8, 18, 24(c), and 25(a) of FIFRA, 7 U.S.C. 136 et seq., as amended, sections 408, 409, and 701 of the Federal Food, Drug and Cosmetic Act (FFDCA), 21 U.S.C. 301 et seq., and the Reorganization Plan No. 3 of 1970.

B. Background

EPA originally published FIFRA GLP standards in the Federal Register of November 29, 1983 (48 FR 53946), which were codified as 40 CFR part 160. At the same time, EPA published GLP standards applicable to testing required under the Toxic Substances Control Act (TSCA, 48 FR 53922, 40 CFR part 792). These regulations were promulgated in response to investigations by EPA and FDA during the mid-1970s which revealed that some studies submitted to the Agencies had not been conducted in accordance with acceptable laboratory practices. Some studies had been conducted so poorly that the resulting data could not be relied upon in EPA's regulatory decision-making process. For instance, some studies had been submitted which did not adhere to specified protocols, were conducted by underqualified personnel and supervisors, or were not adequately monitored by study sponsors. In some cases results were selectively reported, underreported, or fraudulently reported. In addition, it was discovered that some testing facilities displayed poor animal care procedures and inadequate recordkeeping techniques. The FIFRA GLP standards specify minimum practices and procedures which must be followed in order to ensure the quality and integrity of data submitted to EPA in support of a research or marketing permit for a pesticide product.

When EPA published its final FIFRA and TSCA GLP standards in the Federal Register of November 29, 1983, EPA sought to harmonize the requirements and language with those regulations promulgated by FDA in the Federal Register of December 22, 1978 (43 FR 60013), and codified as 21 CFR part 58. Differences between the two Agencies' current GLP regulations exist only to the extent necessary to reflect the Agencies' different statutory responsibilities under TSCA, FIFRA, and the Federal Food, Drug and Cosmetic Act (FFDCA). Similar to the FDA GLP regulations, the FIFRA and TSCA GLP standards delineate standards for studies designed to determine the health effects of a test substance; however, the TSCA GLP standards also contain provisions related to environmental testing (i.e., ecological effects and chemical fate).

Compliance with EPA's FIFRA and TSCA GLP standards has been monitored through a program of laboratory inspections and study audits coordinated between EPA and FDA. Under an Interagency Agreement originated in 1978, FDA carries out GLP inspections at laboratories which conduct health effects testing. EPA primarily performs GLP inspections for environmental laboratories and conducts data audits for health effects and environmental studies.

After a thorough review of its GLP regulations and compliance program, FDA concluded that some of the provisions of the GLP regulations needed to be clarified, amended, or deleted to reduce the regulatory burden on testing facilities. Accordingly, FDA revised its GLP regulations in the Federal Register of September 4, 1987 (52 FR 33768). These GLP revisions are intended to simplify the regulations without compromising study integrity.

EPA agrees with FDA that many provisions of the GLP regulations can be streamlined without compromising the goals of the GLP standards. Therefore, EPA is amending the FIFRA GLP standards to incorporate many of the changes made by FDA to its revised GLP regulations. In addition, EPA is expanding the scope of the FIFRA GLP standards to include the environmental testing provisions currently found in the TSCA GLP standards. EPA's revision to the FIFRA GLP standards also extends their scope to include product performance data (efficacy testing) as currently required to be submitted by 40 CFR 158.640. In summary, the FIFRA GLP standards will allow EPA to ensure the quality and integrity of all data submitted in support of pesticide product research or marketing permits. Elsewhere in this Federal Register, EPA is making similar changes to the TSCA GLP standards.

Return to the Main Quality Assurance Page

It is EPA's policy to minimize the regulatory burden on the public which might arise from conflicting requirements promulgated under different regulatory authorities. In keeping with this policy, the final FIFRA 1983 GLP standards, 40 CFR part 160, followed the format and with few exceptions, the wording of FDA's final GLP regulations, 21 CFR part 58. Differences between the EPA and FDA GLP regulations were based upon varying needs and responsibilities under each Agency's regulatory statutes. This revision to the FIFRA GLP standards follows this same policy by conforming to many of the changes FDA made to its GLP regulations, published in the Federal Register of September 4, 1987 (52 FR 33768). EPA has varied from FDA's revised GLP regulations only when necessary due to EPA's statutory responsibilities. The most significant differences between the EPA and FDA revised GLP regulations are the scope of the testing and the test systems affected.

More specifically, EPA is requiring compliance with the FIFRA GLP standards for all studies submitted to EPA which are intended to support pesticide product research or marketing permits. Under the 1983 FIFRA GLP regulations EPA only required GLP compliance under FIFRA for health effects testing. However, unlike FDA, testing required by EPA in support of research or marketing permits may include ecological effects, environmental and chemical fate, and efficacy (as stipulated by 40 CFR 158.640 Product performance data requirements), as well as health effects testing. Therefore, in an effort to attain consistency in the quality and the integrity of all data submitted to the Agency, EPA has determined that it is necessary to expand the scope of the FIFRA GLP standards to require that all types of testing which are used to obtain data in support of research or marketing permits be conducted in accordance with the amended GLP standards that are required to be submitted under 40 CFR 158.640.

EPA's amended FIFRA GLP standards also vary from FDA's in their coverage of testing conducted in the field. To ensure the quality and integrity of all data submitted in support of research or marketing permits, EPA believes that GLP standards must apply whenever data collection occurs. Because many of the test data required by EPA under FIFRA are developed in the field, or more accurately in outdoor laboratories (i.e., ground water studies, air monitoring studies, degradation in soil, etc.), EPA is including field testing within the scope of the standards.

EPA's FIFRA GLP standards also differ from FDA's in the scope of the requirements provided for test system care facilities, test system supply facilities, and test system care. Because testing required by FDA is focused on health testing, in which animals are the central test system, it is appropriate for FDA's GLP regulations to focus on requirements for appropriate animal care facilities (21 CFR 58.43), adequate animal supply facilities (21 CFR 58.45), and proper animal care (21 CFR 58.90). However, the broad range of testing required by EPA may involve plants, soils, microorganisms, as well as animals, for the primary test systems. To ensure the quality and integrity of all data submitted to EPA, 160.43 Animal care facilities, 160.45 Animal supply facilities, and 160.90 Animal care are being expanded to cover facilities, handling, and care of all test systems. Accordingly, EPA is retitling these sections as follows: 160.43 Test system care facilities, 160.45 Test system supply facilities, and 160.90 Animal and other test system care. Further, in most instances, EPA is replacing the term "animal," which is currently used in the FIFRA GLP standards, with the broader term "test system." Specifically, this change occurs in 160.43, 160.45, 160.81, 160.90 and 160.120. These changes are further discussed in Unit II of this preamble.

The remaining differences between EPA and FDA GLP regulations are described in the preamble to this final rule and the preamble to the FIFRA GLP standards, published in the Federal Register of November 29, 1983 (48 FR 53946). EPA has coordinated this final rule with FDA and has considered public comments on the December 28, 1987 EPA proposal (52 FR 48920).

Return to the Main Quality Assurance Page

D. Publication of the Complete Rule

The entire FIFRA GLP rule (40 CFR part 160) is published in this notice to simplify interpretation and facilitate the use of this notice by the regulated community. The following lists the sections of 40 CFR part 160 that were changed from the 1983 rule:

Sections affected Changes

160.3 "Batch," "Control substance," "Study," and "Test system," revised; "Test substance or mixture," removed; "Carrier," "Experimental start date," "Experimental termination date," "Reference substance," "Study completion date," "Study initiation date," "Test substance," and "Vehicle," added.

160.29 (d), (e), and (f) revised.

160.31 (b) and (d), revised.

160.35 (a), (b) (1), and (3), revised; (e), removed.

160.41 Revised.

160.43 Revised.

160.45 Revised.

160.47 Revised.

160.49 Revised.

160.53 Removed.

160.61 Revised.

160.63 (b), revised.

160.81 (b) (1), (2), (3), (5), (6), (7), and (12) and (c), revised.

160.90 Revised.

Subpart F Heading revised.

160.105 Revised.

160.107 Heading and introductory text, revised.

160.113 Revised.

160.120 (a), revised.

160.130 (d) and (e), revised.

160.135 Added.

160.185 (a) (4) and (5), and (c), revised.

160.190 (a) and (e), revised.

160.195 (c), revised; (i) added.

Return to the Main Quality Assurance Page

II. SUMMARY OF COMMENTS AND RESPONSES

EPA received 43 comment letters: 24 from manufacturers of pesticide products regulated by EPA, 8 from associations, 10 from testing or consulting laboratories, and 1 from another government agency. The majority of the comments supported the proposed changes, although numerous suggestions were made for additional revisions to parts of the 1983 FIFRA GLP regulations not subject to this rulemaking or modifications to the proposed changes. Comments that raised important policy questions, suggested modification to the essence of the proposed regulation, or required an individual response, are discussed below. Comments addressing changes to the GLP standards that were not proposed are not the subject of this rulemaking. However, all comments made have been placed in the public record.

A. General Provisions

1. Scope

Comment: EPA should specify exactly what categories of studies (especially efficacy) are covered under the revised GLP regulations since they are discussed in the preamble and will not appear at 40 CFR part 160 when the final rule is published.

Response: EPA intends GLP standards to cover all types of studies required to be submitted and does not feel it necessary to list each type.

Please note that EPA is developing additional product performance regulations. EPA plans to consider the impact that GLP standards will have on these new product performance requirements to determine if the full scope of GLP standards should apply to studies performed to fulfil these requirements. Unless the GLP rule is modified to specifically exclude certain parts of product performance regulations, the full GLP rule will apply to all existing and prospective product performance studies required under 40 CFR 158.640.

2. Definitions

(a) Batch - The definition of "batch" is expanded to include reference substances. This was an omission in the proposed rule that is corrected in the final rule to maintain consistency with the use of the term in 160.105(a).

(b) Carrier

i) Comment: The word "systems" should replace the word "organisms" in the definition of "carrier," to be consistent with the term "test system."

Response: EPA concurs with the suggestion. To be consistent with the definition of "test systems, ' the word is changed accordingly.

ii) Comment: EPA should revise the list in parentheses that follows the word "material" in the definition of "carrier" to make it all inclusive.

Response: EPA has decided to add the phrase "including but not limited to * * *", to indicate that the list provides examples and is not meant to be all inclusive.

(c) Control substance

i) Comment: Since "material" conveys a broader description than "substance" and is already used in definitions for "carrier," "control substance," and "reference substance," "chemical substance" should be changed to "chemical material" in the definition of "control substance."

Response: EPA does not believe that a change in terminology is needed to broaden the definition since the term "material" is already included in the present definition. The term "substance" must be retained to maintain consistency with TSCA and the TSCA GLP standards.

ii) Comment: EPA should delete the phrase "for no-effect levels" in the definition of control substance. The definition as written is too narrow and excludes analytical chemistry (e.g., chemical fate, residues chemistry) operations where the term "control" has a meaning distinctly different from biological effects.

Response: Since the purpose of the analytical control is to establish eventually that none of the materials administered to the test system interfere with identification of the test substance and its degradate(s) and metabolite(s), EPA agrees that the terminology is too limiting and is replacing the phrase "for no-effect levels" with the phrase "for known chemical or biological measurements." The definition now reads: "Control substance means any chemical substance or mixture, or any other material other than a test substance, feed, or water, that is administered to the test system in the course of a study for the purpose of establishing a basis for comparison with the test substance for known chemical or biological measurements."

(d) Experimental start and termination dates

Comment: These dates would be difficult to predict, especially for field studies, because they would be subject to natural or man-made conditions that cannot be controlled or anticipated. Since the dates would be subject to change, many protocol amendments would be required, thereby creating an undue administrative burden.

Response: The experimental start and termination dates specified in the protocol are merely proposed dates. Therefore if the actual experimental start or termination date is different from the proposed dates no protocol amendment shall be required.

(e) Reference substance

Comment: If EPA intended the term "reference substance" to include analytical and calibration standards, then several other sections of the proposed rule which mention "reference substance," would also require the same types of records to be kept for analytical standards. This would constitute an excessive burden on management which would require maintaining various records that do not add any value to the study.

Response: The definition of reference substance is intended to include analytical reference standards. Therefore, EPA has modified the definition of "reference substance," as follows: "Reference substance means any chemical substance or mixture, analytical reference standard, or material, other than a test substance, feed, or water, that is administered to, or used, in analyzing the test system in the course of a study for purposes of establishing a basis for comparison with the test substance for known chemical or biological measurements." EPA believes this change eliminates any ambiguity in the definition.

EPA disagrees that inclusion of analytical reference standards in this part constitutes an excess documentation burden or adds no value to the study. Documentation which supports defining analytical reference standards should not require excessive paperwork since common laboratory practices already require assurance of the validity of standards in order to make certain that the measurements are accurate.

(f) Study

i) Comment: "Basic exploratory studies" are excluded from GLP standards, but the results of such studies may be required to meet GLP standards, if included in support of research or marketing permits.

Response: EPA does not wish to discourage basic exploratory testing and does not explicitly require GLP standards for such tests even if the data are later submitted to EPA. However, if the data are to be used in sole support of a marketing permit such non-GLP studies may not be accepted. GLP standards are required when data is developed in the context of a study that is required to be submitted to EPA in support of a research or marketing permit. When GLP standards were not followed in the case of a study performed with the original intent of exploratory testing, a GLP compliance statement should be included in the study report to indicate this.

ii) Comment: It is not clear what constitutes separate studies and what studies could be included under a single protocol. Specifically, is a test system located in several different geographical locations a single study or would each location by means of its particular requirements need to be a separate study?

Response: The protocol defines what the study entails. Therefore, if the test system for a specific study is located in different geographical locations, the protocol will describe the study as being located at the different sites. EPA is adding the phrase "at one or more test sites" to the definition of "study" to clarify the intent that more than one field site may be included in one study.

iii) Comment: The proposed definition of study would imply that each determination such as stability, solubility, octanol water partition coefficient, volatility, persistence, and other data point determinations would be separate studies with concomitant requirements such as protocols and quality assurance unit (QAU) inspections.

Response: EPA intends that QAU inspections as listed in 160.35 be conducted at intervals adequate to ensure the integrity of the study for each determination such as stability, solubility, octanol water partition coefficient, volatility, persistence, and other data point determinations. However, if done as part of a larger study, then these determinations are covered under the larger study's protocol or standard operating procedure (SOP). If they are submitted to EPA as studies unto themselves, then they do require their own protocols.

iv) Comment: An experiment such as product chemistry which does not involve a test system cannot be considered a "study" and therefore would not be covered by GLP standards.

Response: Studies designed to determine the physical or chemical characteristics of a test substance are included within the scope of these regulations. Therefore, EPA intends to include product chemistry experiments in the definition of "study." This change is consistent with the definition of the term "study" as it now appears, and as it appears in the TSCA GLP standards at 40 CFR Part 792. In the case of product chemistry experiments, the test substance itself may be the test system.

v) Comment: The addition of the term "or in the environment" to the definition of "study" indicates that the change extends the proposed regulations to field studies. While it is necessary to ensure the validity of all data collected, the variety and special requirements of field research have not been addressed in the new rules

Response: These regulations are intended to apply to all studies required to be submitted under FIFRA, including those conducted in the field. EPA recognizes that field studies vary and have special requirements, but believes that the development of protocols and SOPs by the testing facility provides adequate flexibility in this respect.

vi) Comment: Why are metabolism, product performance, environmental and chemical fate, persistence and residue listed in the definition of "study", but not toxicology data or data to assess hazards and product chemistry.

Response: The list is not meant to be limiting in any way. Data to assess toxicology, hazards and product chemistry are included under "effects" and "other characteristics" under the new definition of "study".

vii) Comment: "Prospectively" should not be deleted from the definition of study. If the essence of GLPs requires a carefully planned study and the proposed rule is very strict about documentation that must be completed prior to the experimental start date, how can the GLP standards also apply to studies that were generated without a protocol or advance planning, such as epidemiology.

Response: EPA disagrees with the comment. The term prospectively is deleted because EPA wishes all studies including epidemiological studies where past exposure to a study population is determined or estimated retrospectively, to be performed under GLP standards. EPA recognizes that in such studies data used may not have been generated in conformance with FIFRA GLP standards. However, it is EPA's position that the epidemiological study itself can be conducted and submitted to EPA in accordance with the GLP standards. Retrospective aspects of such studies that are not performed according to GLP standards, for example, test system treatment, should be identified in the compliance statement submitted with the study report.

In addition, the types of studies potentially not covered by these regulations were expanded in the definition of "study" to include experiments involving test methods.

(g) Study initiation and completion date

Comment: EPA should delete the definition of "study initiation date" and "study completion date," since these terms were not defined in the 1983 GLP standards. The dates will be included in the protocol and final report and do not need further emphasis.

Response: EPA believes that it is necessary to define the terms to differentiate them from "experimental start and termination" dates. These terms indicate the dates on which specific milestones occur during a study. The definition is necessary to clarify EPA's requirements, and to ensure consistency with FDA's GLP regulations (52 FR 33780).

The phrase "close of the study" as used in 160.33(f), and the phrase "study is completed" as used in 160.195(b)(3) both refer to the "study completion date." Therefore, as of the study completion date: (1) Under 160.33(f), the study director must ensure that all raw data, documentation, protocols, specimens, and final reports are transferred to the archives; (2) after this date under 160.185(c), corrections or additions to the final report must be in the form of an amendment by the study director under the procedures specified in that section; and (3) in the applicable situations described in 160.195(b)(3), records must be maintained for a period of at least 2 years following the study completion date.

Furthermore, the phrase "study is initiated" as used in 160.31(a), and the phrase "study was initiated" as used in 160.35(b)(1) would refer to the "study initiation date." Therefore, as of the study initiation date: (1) Under 160.31(a), the testing facility management would designate a study director; (2) under 160.35(b)(1), the study would be entered on the master schedule sheet by the QAU; and (3) under 160.120(b), after this date all changes or revisions in the protocol would be documented, signed by the study director, and dated. EPA also expects that as of the study initiation date, under 160.31(e), the testing facility management would have ensured that personnel, resources, facilities, equipment, material, and methodologies are available as scheduled.

(h) Test system

Comment: What constitutes the "test system" in tests of pre-emergent herbicides, soil pesticides, and product chemistry studies?

Response: The definition of "test system" includes the statement that it is "* * * any * * * chemical or physical matrix * * *", including subparts thereof that are treated with the test, control, or reference substance and also appropriate components of the system that are not treated. Therefore, test systems may include the soils that pesticides are applied to, and in the case of product chemistry, the test system may be the test substance itself.

EPA is including the term "reference," which was inadvertently omitted from the definition as it appeared in the proposed rule. In addition, EPA is replacing "e.g." in the parenthetical with "including but not limited to" in order to clarify that it is not our intent for the list to be all encompassing.

(i) Vehicle

Comment: The definition of "vehicle" serves to clarify the GLP standards, but there has been no confusion based on the current standards and this change is contrary to EPA's stated objective of being consistent with FDA's GLP regulations.

Response: EPA believes that clarification is needed. The EPA GLP standards cover a larger number of types of studies and the need for clarification of the meaning of potentially ambiguous terms is greater.

Return to the Main Quality Assurance Page

B. Organization and Personnel

1. Testing Facility Management

Comment: The specific requirement to document the replacement of the study director as raw data should be retained. The "master schedule" should not be considered "raw data" as was indicated in the preamble (52 FR 48923) to the proposed rule.

Response: EPA deleted the requirement that the replacement of a study director must be documented as "raw data" to conform to the revised FDA GLP regulations. This is because replacement of the study director must be reflected on the master schedule sheet, which is a study record that must be retained.

In addition, the term "reference" which was inadvertently omitted in the proposed rule, has been added to 160.31(d).

2. Study Director

Comment: Archiving the study records within a "reasonable period" after the study completion date, instead of at the close of he study as required by 160.33(f) would not impact on the integrity of the records.

Response: EPA believes that the requirement that all raw data, documentation, protocols, specimens, and final reports be transferred to the archives at a definitive time, i.e., the study completion date, is necessary. This assures an intact audit trail for the study.

3. Quality assurance unit

i) Comment: A QAU that is entirely separate from and independent of the personnel engaged in the conduct of the study creates an unjustified financial burden on some facilities. In some cases it would be impossible to establish a completely independent QAU with qualified personnel.

Response: As stated in the proposed rule (52 FR 48920), EPA does not require the QAU to be a fixed, permanently staffed unit whose only functions are to monitor the quality of a study. EPA is only concerned that there be a distinct separation of duties between those personnel involved with the conduct or direction of a study and those personnel performing quality assurance on the same study. Therefore, 160.35(a) prohibits personnel from performing quality assurance activities on their own study. The regulations allow a study director for a particular study to serve as a part of the QAU or as the QAU for a different study. FDA noted (52 FR 33771) that it was aware that many small laboratories could not afford the operation of a permanently staffed QAU. EPA would like to point out that in those situations where there are different individuals performing the quality assurance functions for different studies, each individual is required to maintain that portion of the master schedule sheet which relates to the study being monitored. For this reason, EPA agrees with the FDA's conclusion that the separation of functions on a study-by-study basis, as permitted in the existing and revised regulations, would provide effective quality assurance. In view of the potential gain to management, to sponsors, and to EPA, through the added assurance of well-conducted studies, the increased costs are thereby justified. EPA believes that its intent is more clearly indicated by the changes now being made.

ii) Comment: EPA should delete the requirement to index the master schedule by test substance, and the QAU should only be required to index the master schedule to facilitate retrieval of the information monitored.

Response: EPA acknowledges that a test facility may have several studies in progress on each test substance that is listed on the master schedule sheet. However, EPA concludes that deleting the requirement to index by test substance would be inappropriate, since the master schedule sheet is the mechanism through which the QAU can assure management that the facilities are satisfactory and there are adequate numbers of competent personnel available to perform the scheduled tasks. Furthermore, 160.31(e) requires that management assure that study materials (e.g., test substances) are available as planned. Therefore, elimination of this requirement would hinder a major function of the master schedule sheet and hamper the conduct of a critical management role.

iii) Comment: Laboratory management should have the discretion to determine who enters the data into the master schedule, as long as the required information is listed.

Response: EPA believes that management retains such discretion since it is involved in determining the composition of the QAU and it provides an adequate number of such personnel ( 160.31(c) and (e)). The QAU is distinguished by training that ensures that QAU functions are properly conducted. As stated above, study personnel may belong to the QAU as long as they are not performing the QAU functions associated with studies they are involved in.

iv) Comment: Do all studies conducted by an analytical laboratory have to be listed on a master schedule, or just those studies that will be, or likely be, submitted to EPA?

Response: The GLP standards specifically exempt many product chemistry studies as described in 160.135. The master schedule need only list those analytical chemistry studies that will be or will likely be submitted to EPA.

v) Comment: The requirement for inspection of each study under 160.35(b)(3) regardless of duration is excessive for the quality assurance needed to address study integrity, especially where studies are performed by highly standardized procedures. The repetitive inspection of these types of studies would consume large amounts of time for both the study personnel and QAU staff. Auditing each study is not necessary to ensure the work is conducted in compliance with the regulations. Random sampling procedures should be allowed in selecting studies and phases of studies to inspect to decrease the work load and resource requirements of the QAU.

Response: EPA does not believe that a random inspection program would be an appropriate method of evaluating a study. Generally, random sampling provides an adequate means of quality control where analysis involves repetition or identical procedures. However, any assumption that the conduct of one phase of one study would be representative of another would be invalidated by the differences among study personnel and the operations they conduct. Furthermore, this requirement is not intended for all routine studies. Section 160.35(b) is among the exclusions for chemical and physical characterization studies as listed in 160.135(b).

In conformance with the revised FDA GLP regulations (52 FR 33780), EPA modified the requirements of 160.35(b)(3) to provide for inspections of a study on a schedule adequate to ensure the integrity of the study. The changes to this section will allow the QAU the necessary latitude to adjust its monitoring activities to meet the individual needs of each study. However, each study, no matter how short, must be inspected at least once while in progress. EPA expects that by allowing the QAU flexibility in designing a reasonable inspection schedule, the goal of ensuring the quality of the study can be best achieved.

vi) Comment: EPA indicates in the preamble to the proposed rule ( 160.35(e), (52 FR 48923)) that all QAU records will now be routinely available to inspectors. Existing GLP standards treat certain QAU records as confidential, and explicitly state that the only QAU records to be reviewed by EPA auditors would be the master schedule (e.g., the inspection dates, study inspected, the phase or segment of the study inspected and the name of the individual performing the inspection). If QAU records for findings and corrective action are available on an auditor's request, QAUs would lose their effectiveness.

Response: EPA shares the concerns of the commenters that access to all parts of a QAU inspection would weaken the inspection system, and recognizes the need to maintain a degree of confidentiality. Therefore, records of findings and problems, as well as records of corrective action recommended and taken, are exempt from routine EPA inspections, except under special circumstances as indicated in 160.15. However, EPA maintains that all other reports and records must be easily accessible and made available to EPA and FDA inspectors when requested as indicated in 160.35(c).

Return to the Main Quality Assurance Page

C. Facilities

1. General

i) Comment: Outdoor testing facilities should not be under GLP standards since: (a) Outdoor test facilities will be conducting studies according to approved protocols; (b) ensuring suitability is highly subjective based on the diverse number of possible locations; (c) there is a concomitant lack of clear standards for determining suitability of locations. Procedures must be specified by EPA regarding the determination of suitability for locations, testing facilities, etc.

Despite best efforts, the choice could always be subject to criticism and even criminal liability based on a good faith Compliance Statement indicating GLP standards had been followed. Most outdoor testing is done to mimic normal agricultural conditions which are specific for the test substance and use being proposed. Therefore, the determination of whether the size, construction or location of a facility is suitable for a study is a technical issue, and is not within the scope of the GLP regulation and would be considered in the experimental design of the protocol.

Response: In cases where an EPA-approved protocol establishes test locations, that protocol would satisfy GLP requirements. EPA considers any site to be the testing facility wherever testing is undertaken to generate data required to be submitted to EPA. The conditions required by the protocol are not necessarily conducive to artificial manipulation in the field, or to other outdoor testing facilities. Therefore, ensuring the suitability of the location of these types of testing facilities is both a valid and necessary part of protocols approved by EPA.

ii) Comment: The design of the individual scientist could be dictated by 160.41 since a "testing facility" (definition from 160.3) means "a person who actually conducts a study * * *" The term "test site" should be defined to refer to the actual location of a given "test system." "Testing facility" could then be used as currently defined and refer to an individual (mobile development scientist or scientist working from a testing farm facility).

Response: The definition of "person" in this Part refers to the legal entity responsible for testing, including organizational units. Consequently, it does not specifically indicate an individual scientist.

2. Test system care facilities

i) Comment: Instead of expanding the original document to fit all test systems, the old rules should be left as is, and a statement added to cover non-animal test systems.

Response: EPA disagrees with the comment and believes that specific changes of the old rule are necessary to avoid ambiguity concerning the meaning of non-animal systems.

ii) Comment: Section 160.43(a)(2) and (b), (e), (f), (g) and (h) should be deleted because EPA has already stated that these GLP requirements will be applicable to all types of testing. It is not necessary to add the four new paragraphs detailing specific requirements of environmental conditions for aquatic organisms and plants.

Response: EPA believes that some test systems, e.g. aquatic, are unique, and for the sake of clarity, they require special treatment in the regulations.

iii) Comment: Field studies should be exempted because isolation is not possible in these types of studies.

Response: EPA disagrees and believes that inclusion of field studies poses no unusual burden, since the separation is only required to be "as needed" to ensure "proper separation." If the procedures used are justifiable based on experimental design and documented, then this requirement is met. "Proper separation" in a field study may mean simply that only one crop is planted in the same subplot.

iv) Comment: The change in 160.43(c) is appropriate but the current wording does not require separate disease handling facilities in every case. The proposed change has merit in clarifying the options available to laboratories and the change promotes harmony between EPA and FDA GLP regulations.

Response: EPA agrees with the comment. In 160.43(c), EPA is deleting the requirement that separate areas be provided in all cases for the diagnosis, treatment, and control of test system diseases. Instead, a change is made so that separate areas are provided "as appropriate." This change is consistent with the September 4, 1987, revised FDA GLP regulations and the revised TSCA GLP regulations.

EPA has made this change to allow laboratories the option of disposing of diseased test systems without also bearing the expense of maintaining separate areas in testing facilities for diagnosis, treatment, and control of disease. Additionally, EPA recognizes that the diagnosis and treatment requirements of 160.43(c) may not be appropriate when dealing with such test systems as soil, plants, or microorganisms. However, if the decision is made not to dispose of the test system, test system care facilities, as specified in 160.43(c), must be provided.

3. Test system supply facilities

i) Comment: The first sentence in 160.45(a) should be changed so that plants and plant materials are covered in this section.

Response: EPA believes that since plants and plant materials are covered in 160.45(b), including them in 160.45(a) in unnecessary.

ii) Comment: Change 160.45(b) by deleting it or expanding it to include tests not confined to the indoor laboratory or greenhouse.

Response: EPA agrees with the comment and is expanding the wording of 160.45 to emphasize that this section is not intended to be confining. Therefore, 160.45(a) is changed to read "* * * areas where the test systems are located * * *," and 160.45(b) is changed to read "* * * (included but not limited to fields, greenhouses,* * *)."

iii) Comment: The addition of the two new paragraphs outlining plant and aquatic facilities to 160.45(b) is unnecessary. These considerations are addressed in 160.41 with the requirement that testing facilities be of suitable construction "to facilitate proper conduct of studies."

Response: EPA maintains that testing facilities as mentioned in 160.41 and test system supply facilities as mentioned in 160.45, are not the same and must be addressed separately.

iv) Comment: EPA should delete 160.45(b) introductory text, (b)(1), (b)(2), and (c) because this information was adequately covered in 160.45(a) and in 160.43, and the facilities they refer to will be addressed in study protocol.

Response: EPA maintains that 160.43 (test system care) is different from 160.45 (test system supply facilities) and must therefore be treated separately.

4. Facilities for handling test, control, and reference substances

i) Comment: These requirements would severely restrict the ability of efficacy investigators to test their product, since 160.47 would require separation of facilities for test animals and testing material. The real issue for efficacy testing is test substance accountability which should be a vital part of the efficacy testing protocol, and appropriate records maintained to verify test substance accountability.

Response: EPA notes that similar concerns were raised by commenters regarding the 1983 rule. The wording "as necessary" was included then to allow latitude in facility design and operation. EPA agrees that other measures, i.e. protocol, SOPs, and appropriate records, must be adequate to demonstrate the integrity of test, control, and reference substances during handling.

ii) Comment: Would it be necessary to provide separate sink facilities or separate rooms for mixing of the test, control, and reference substances or for adding water to tank sprayers?

Response: Separate areas are required for receipt, mixing and storage of test, control, and reference substances and their mixtures as necessary to prevent contamination or mixups. The same sink could be used for all work involving mixing provided that the procedures (SOPs) used are adequate to prevent contamination and mixups. Separate areas for receipt and storage and for mixing and storage of test, control, and reference substances as required in 160.47(a)(1), (2), and (3) does not mandate the use of separate rooms. The areas could be in the same room provided there is adequate space and equipment to provide that contamination and mixup do not occur. This determination should be made on a case-by-case basis.

Return to the Main Quality Assurance Page

D. Equipment

1. Maintenance and calibration of equipment

i) Comment: The entire section 160.63(b), requires unnecessary documentation and/or is vague about what is required, especially for field portions of residue studies. Equipment used in these studies may only be used on an occasional basis, and routine inspection should only be "before use." Requiring calibration and maintenance logs for all equipment involved in generating a residue sample would be prohibitive, would often be forgotten or overlooked and would then be a cause for not meeting audits.

Response: The requirement states that equipment shall be "adequately inspected, cleaned and maintained" and "adequately tested, calibrated, and/or standardized." This requirement is not changed from the old rule. The laboratory has latitude in defining in its SOPs what is "adequate" unless given specific guidance otherwise (i.e. in test rules or testing guidelines). However, EPA recommends that calibration and maintenance records be available for all equipment used in field studies. This includes equipment used only rarely and rental equipment.

ii) Comment: It is better to designate in 160.63(b) that repair and maintenance will be performed by "qualified personnel," than to require that a person be designated in the written SOP. The requirement for written SOPs in 160.63(b) causes problems since at many laboratories the equipment used in conducting a study is shared by a number of individuals and the care and maintenance of the equipment is also shared. In the event of equipment failure, a number of laboratory personnel may be capable of repairing or correcting a problem, or in more serious equipment failures, a service representative of the manufacturer may be called. It is therefore difficult and very inefficient to designate specific people to perform each specific maintenance and repair operation.

Response: The definition of "person" as it appears in 160.3(h) is not limited to an individual scientist or technician, but includes an organizational subunit. Consequently, the SOP that designates the "person responsible" will be designating a subunit of the testing facility, which could be one or several individuals. This view is consistent with FDA's (52 FR 33774) interpretation and definition of "person." Where duties are delegated in the SOPs, all contingencies may be addressed, including the contracting of service personnel.

iii) Comment: Certain pieces of equipment such as tractors, land preparation and land measuring devices should be exempt from the calibration requirement, as should standard commercially available laboratory ware, such as graduated cylinders, beakers, flasks, etc. Only equipment directly related to application of the test substance, such as sprayers or granular applicators should be listed as requiring calibration. Therefore, 160.63(c) is not appropriate for field studies.

Response: EPA believes that calibration should be required for the application phase of field studies. However, the method of calibration, and hence the exact equipment to be calibrated, are not specified in GLP standards, as long as the methods and records ensure the quality and integrity of the study. Some equipment, such as graduated cylinders and volumetric flasks are pre-calibrated and do not need to be recalibrated. Equipment directly related to the application of the test substance may require calibration, but application rates may include other parameters. The methods used to measure all parameters inherent in the determination of application rates would have to be adequately calibrated in order to ensure the quality and integrity of the study.

Return to the Main Quality Assurance Page

E. Testing Facilities Operation

1. Standard operating procedures

i) Comment: There are few standardized tests available to researchers related to novel microbial pesticides. An experimental use permit is required for the evaluation of certain microbials at an earlier stage of research than is required for chemical evaluations. Therefore, it would be very cumbersome to require written SOPs for microbial pesticides, since the methodology may be in a state of flux. It may only be possible to develop SOPs following the completion of a study. If methods of application and assessment need to be modified for each microbial developed, it would be best to affirm that methods development could be performed in accordance with accepted scientific standards without having SOPs as described in 160.81. EPA is encouraged to take a flexible, case-by-case approach to establishing appropriate GLP standards for a given set of experiments concerning development of microbial pesticides. Allowances could be made for situations in which SOPs are inappropriate, such as in the early stages of field work. These allowances made in advance of the work, could then be positively affirmed as good laboratory practice, rather than as tolerated, non-compliance with GLP standards. This would alleviate the uncertainty of performing experiments in a scientifically sound fashion, without knowing until the conclusion of the work whether the data would be acceptable to EPA.

Response: EPA agrees that there are special problems associated with the early stage of method development. Method development phases of an experiment are not under GLP standards as has been clarified in the definition of "study" in 160.3. SOPs are thus required for those operations in which all steps have been worked out. However, SOPs are needed to ensure the quality and integrity of all studies performed under GLP standards, for instance, after the method has been developed. There is flexibility in relation to SOPs insofar as changes can occur during the study as long as they are authorized by the study director (and management, if the changes are significant) and documented with raw data. Furthermore, methodology that is not generalized or established sufficiently to be included in SOPs can be defined in the study protocol.

ii) Comment: Although unchanged from the old rule, the second half of 160.81(a) should not apply in some cases. The justification for this is as follows: (a) Unforeseen circumstances cannot be authorized; (b) minor deviations do not need authorization by the study director; (c) people who conduct the studies are required to be appropriately trained and are able to make decisions if necessary to deviate from the SOPs; (d) in field studies, deviations from SOPs will occur before the researcher is able to consult with the study director; (e) decisions about deviations from SOPs that are made by field personnel would be based on standard agricultural practices.

Response: EPA disagrees with the suggestion that some deviations do not require authorization by the study director. It is necessary for the study director to authorize deviations from SOPs to ensure that these deviations do not have an adverse impact on the study. SOPs should be written with sufficient flexibility to accommodate field studies by anticipating conditions under which appropriate actions must be taken without the need for authorization by the study director. Standard agricultural practices can be referenced in SOPs as long as this does not lead to ambiguity concerning appropriate action to be taken in a given situation. If SOPs state the constraints on action and a decision is made within these limits, there is no deviation. This is in concert with FDA's GLP regulations (52 FR 33774) which require that the study director make certain that specified procedures are followed and that all modifications to the procedures in the approved study plan are documented and approved.

iii) Comment: Some of the examples of required SOPs provided in 160.81(b) are not applicable to all test systems or study types. For example, "test room preparation" would not be appropriate when conducting field residue trials, and "necropsy of test system or postmortem examination of test systems," would not apply to studies using a chemical or physical matrix as the test system (sterile water, soil, agricultural fields). Furthermore, 160.81(c) states that, "Each laboratory or other study area shall have immediately available manuals and SOPs relative to the laboratory or field procedures being performed."

Response: EPA agrees that the term "room" in 160.81(b)(1) is inappropriate to many studies and is changing the word to "area" in order to clarify that field studies are included. EPA believes that 160.81(b) should apply in all cases since the purpose of SOPs is to insure the quality and integrity of the data generated in the course of a study as stated in 160.81(a). However, procedures that are not necessary to be performed, such as necropsy in the case of field studies, do not require SOPs.

iv) Comment: The term "test systems" should not replace "animals" in 160.81(b)(6) and (7). Although this requirement is useful for preventing or slowing autolysis for toxicology studies, for other studies, such as metabolism, addressing the handling of moribund or dead test systems is not appropriate. In these types of studies, if a test system were moribund or dead, the testing guidelines require the part of the study that was impacted to be repeated, and this requirement is only applicable to animals.

Response: EPA disagrees with the comment. This rule applies to plants as well as animals.

v) Comment: Published literature (e.g., ASTM methods) should be acceptable in 160.81(c) as an appropriate part of an SOP and not just as a supplement to a written SOP. The written SOP could incorporate the published literature by reference, without having to rewrite the entire procedure.

Response: EPA agrees that it would not be appropriate to rewrite published literature, hence the allowance for SOPs to use it as supplements. The SOPs are still needed to establish the relationship of the method to data collection procedures and needs in the laboratory. While the resulting SOP would still have to be written, it would in effect be abbreviated in that all of the methodology referenced would not need to be rewritten.

2. Animal and other test system care

i) Comment: Section 160.90(a) should be deleted since the subject is covered in 160.81(b).

Response: EPA recognizes that 160.81(b) requires testing facilities to establish SOPs for animal and other test system care. Section 160.90(a), however, expressly specifies that SOPs shall also cover test system housing, feeding and handling. This section is consistent with FDA's GLP regulations and is not an additional requirement.

ii) Comment: Section 160.90(b) should be simplified to provide that test systems be evaluated prior to use but not necessarily isolated. For some studies, such as plant metabolism, isolating the plants or soil is not appropriate.

Response: EPA disagrees. Isolation is necessary to insure that a test system is free from disease or other conditions that may impact the study. Further, the inclusion of this is consistent with FDA's GLP regulations.

iii) Comment: The evaluation of certain test systems according to "acceptable * * * scientific practice" creates some difficulty, particularly for plants, microorganisms, soil, and water, since such practices are not defined. "Acceptable" should be deleted regarding scientific practice and the requirement be only that a scientific basis be used in determining appropriateness for testing. In this way, testing facilities would not need to justify or prove their basis to be "acceptable" in ill-defined areas or those in flux.

Response: EPA agrees that the term "acceptable scientific practice" may not be definable when method developments are in flux. The term "acceptable" is retained, but the term "scientific practice" is changed to "scientific methods." This change preserves EPA's intent that rigorous scientific methodology be used without implying that rigid practices be adhered to where they may not appropriately exist.

iv) Comment: The requirement under 160.90(c) that the test area be disease-free prior to study initiation is inappropriate for field studies since it would be impossible to declare areas totally disease free under field conditions. Also, one of the objectives of performing studies in the field is to conduct the studies under representative environmental conditions which includes encountered disease and insect pressures, making this part in direct conflict with the study objective.

Response: The requirement is for the test system to be "free of disease or condition that interfere with the purpose or conduct of the study." The current wording therefore provides sufficient latitude for field studies. Furthermore, EPA does not intend compliance with this provision to require deviation from accepted agricultural practices. If disease and insect pressures are considered to be an integral part of a study, they clearly do not interfere with the purpose and conduct of that study. The test system would therefore not need to be free of them.

v) Comment: Section 160.90(c) should be deleted since the effect of corrective treatment cannot be accounted for in test results.

Response: EPA believes that while the effects of corrective actions taken to isolate and treat disease or signs of disease may complicate interpretation of test results, so might the effects of the disease itself. This requirement for field studies is not inconsistent with its inclusion for laboratory, i.e., toxicology, studies.

vi) Comment: Markings which identify animals individually, rather than the group as required by 160.90(d), are needed in many studies with warm-blooded vertebrates in pens, or in the field. For example, precocial young of avian species should be marked individually.

Response: Specific criteria for marking of individuals to meet study requirements should be addressed separately in the protocol of the study. The requirement in 160.90(d) addresses the need that test systems be adequately identified to prevent confounding with other test systems. Identification of precocial birds, for example, may be outlined in the study protocol.

vii) Comment: The proposed multispecies housing under 160.90(e)(1) is redundant to proposed 160.43(a)(1) and is inconsistent with EPA's desire to streamline GLP standards.

Response: EPA disagrees with the conclusion that these sections are redundant. While 160.43(a)(1) states that the facilities shall be sufficient to allow proper separation of species, 160.90(e)(1) refers specifically to test system care within the facilities.

viii) Comment: Field studies should be exempt from the periodic testing requirement of 160.90(g). A bioassay or chemical analysis prior to study initiation should suffice to show that contaminants are not present at levels capable of interfering with the study. The need for prior analysis may even be obviated by documentation of the previous history of pesticide use in the soil according to Standard Evaluation Procedures to ensure that no interfering contaminants are present.

Response: The regulations as written do not require that periodic tests be performed during a study unless there are "contaminants known to be capable of interfering with the study and reasonably expected to be present at levels above those specified in the protocol." If there is no reasonable expectation that a problem exists, periodic testing is not needed. An acceptable method to determine this, such as evaluation of the history of pesticide use, should be defined in the protocol or SOPs.

ix) Comment: The requirement in 160.90(j) for acclimatization of plants and animals should be deleted, since it is not defined and promotes confusion. Animal toxicology tests would be subject to isolation and separately to acclimatization. Organisms in environmental studies will have been isolated with their health status being evaluated per 160.90(b) and acclimatization would have already been performed as part of the process. This part should be amended to indicate that test organisms be acclimatized to all experimental conditions except the test substance.

Response: EPA believes that the term acclimatization has common meaning that is clear in the context of its usage in the regulation. Acclimatization implies accustoming to experimental, i.e., environmental, conditions other than the actual introduction of the effect (e.g., test substance) to be measured in the experiment. If acclimatization is achieved during the process of isolation, it should be so stated in the protocol and does not require additional technical effort.

In addition, the term "organisms" in 160.90(j) has been changed to "systems." This change is consistent with the intended expansion of GLP standards and was an inadvertent omission in the proposed rule.

Return to the Main Quality Assurance Page

F. Test and Control Substances

1. Test, control, and reference substance characterization

i) Comment: Requiring stability and solubility before testing would result in a costly burden to the efficacy testing sponsor. The solubility testing portion of this requirement would not cause significant problems, but requiring stability testing to be completed before study initiation could result in significant time and cost burdens.

Response: It is more costly to have to repeat a study because of inadequate solubility or stability in respect to experimental needs. EPA agrees, however, that requiring stability testing to be completed before study may result in unnecessary delays and is allowing concurrent stability testing. Therefore, EPA has changed the requirement to allow stability testing concurrently with the study. Solubility, where this is relevant to a study, must still be known before the experimental start date. Please note that the 1983 GLP standards require determination of characteristics which will appropriately define the test or control article before study initiation. Thus solubility determination before a study, where it is relevant to the study and hence an appropriate characteristic, is not a new requirement.

ii) Comment: The term "purity" should be expanded to include radiochemical purity since further definition is needed to encompass metabolism/environmental fate studies conducted with radioactive materials.

Response: Radiochemical purity is covered under "other characteristics which appropriately define the test, control, or reference substance." It is not necessary to specifically list this characteristic.

iii) Comment: What level of analysis constitutes "appropriate" characterization? Is quality control batch analysis sufficient? Is it necessary to fully characterize technical materials to 0.1%?

Response: The details of what "appropriately" defines the test substance is a guideline or protocol issue that cannot be