\._.C. N. IS?/.3.‘ I 9-7‘ 87-413 SPR CRS REPORT FOR CONGRESS ! % i .{§ ‘ Vifiig‘ U? E, N IJ B R A FLY ' 2 ‘M; ;‘I:I‘*:.>::-: U N w ERSITY I . L. PESTICIDE MONITORING PROGRAM: DEVELOPING NEW METHODS TO DETECT PESTICIDE RESIDUES IN FOOD T Umt AUG 1 7 I994 Washm t ~ . , Stg Irc:u‘Umvers'ty Ubfaries ‘ '5» M0 63130 Sarah E. Taylor Analyst in Life Sciences Science Policy Research Division April 24, 1987 : - - - Universit of Missouri-Columbi ° ° ’ IIIIHJI |||\fl|JJJ||||| Nlll llllllflll 010- 03939356 3 . ,r. » V . S _ y .u. .. The Congressional Research Service works exclusively for the Congress, conducting research, analyzing legislation, and providing information at the request of committees, Mem- bers, and their staffs. The Service makes such research available, without parti- san bias, in many forms including studies, reports, compila- tions, digests, and background briefings. Upon request, CRS assists committees in analyzing legislative proposals and issues, and in assessing the possible effects of these proposals and their alternatives. The Service’s senior specialists and subject analysts are also available for personal consultations in their respective fields of expertise. ABSTRACT During the past 10 years, several reports have addressed issues concerning the analytical methods used in the Food and Drug Administrations's (FDA) Pesticide Monitoring Program (PMP). In the most recent of these, the General Accounting Office reported, in 1986, that there was a gap between the number of pesticides that could potentially be found in food, and the number that could be detected by laboratory tests FDA generally relies on in its monitoring program. This has raised such questions as, how significant is the gap from a public health standpoint? Why are there not more practicable methods for analyzing pesticide residues in food? Also, what are the potential applica- tions of "new" rapid test methods? CONTENTS ABSTRACTOOIOOO0000000000OIOOOOOOOOIOOOOOOO0000000.0000000000IOOOOOOOOOOOOOOOi SUWARYO000000000000OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOCOOOOOOOOOOOOOOOOOOii INTRODUCTIONOOOOOOO'OOOOIO00000000000000OOOOOOCOOOOOOOCOOOOOOOOOOOOOOOIOI BACKGROUNDOOOOOOI000000OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO2 HISTORY OF RECOMMENDATIONS CONCERNING DEVELOPMENT OF METHODS!DQ0000000000O0OOIOOOOICCOCOCCOOOOOOOOI0000000000O0CO7 FDA's CURRENT RESEARCH PRIORITIES IN METHODS DEVELOPMENT...............10 CURRENT ISSUES IN METHODS DEVELOPMENT..................................12 Multiresidue methods..............................................l3 What Are Multiresidue Methods?...............................13 why Are There Not More Multiresidue Methods?.................16 Research Activities..........................................17 Limitations of Multiresidue Methods..........................21 Rapid Test Methods................................................22 Enzyme Inhibition Method.....................................23 Immunoassay Methods..........................................26 Research Activities..........................................3l CONCIJUSIONIOOO0OOOOIOOOIOIOOOOOOOIOOOOOOOOOOOIOOOOOOOOOOOOOOI000000000033 SUMMARY During the past 10 years, several reports have addressed issues concerning the analytical methods used in the Food and Drug Administration's (FDA) Pesticide Monitoring Program (PMP). In the most recent of these, the General Accounting Office, in 1986, reported that there was a significant gap between the number of pesticides that could potentially be found in food, and the number that the Food and Drug Administration (FDA) is able to detect practicably with classical laboratory methods now in use. As a consequence, a number of potential pesticide residues in foods are not regularly monitored, including several FDA considers to be of high priority, based on potential health risk. These reports have raised anew longstanding issues regarding the significance of the gap from a public health standpoint, and issues related to the need for additional practicable analytical methods. Several "new" analytical methods that make use of the biological reagents, appear to potentially complement the capabilities of classical methods. How- ever, most work to apply these methods to pesticide analysis has been done by university researchers and a handful of private firms. Governmental agencies in general have been minimally involved, and FDA has sponsored no research on these methods. Some observers believe such a Federal commitment is necessary to spur appropriate commercialization in rapid test methods. Whether these methods can be incorporated beneficially into FDA's monitoring program depends, in large part, on whether they can be validated (proved valid and reliable). If these methods are adopted, they present significant implications for the cost and design of the PMP which have not been explored. INTRODUCTION Pesticide analytical methods form the backbone of the Food and Drug Administration (FDA) Pesticide Monitoring Program (PMP). Pesticide analytical methods are scientific techniques used to identify and quantify pesticide residues. The capabilities and limitations of existing techniques have been key determinants of the scope and limits of the FDA program. FDA, which is part of the Department of Health and Human Services (DHHS), is responsible for monitoring all domestic and imported foods for pesticide residues, except for meat, poultry and egg products (which are monitored by the Department of Agriculture). The program has two primary objectives: to enforce pesticide residue tolerances (maximum residue limits for food) established by the Environmental Protection Agency (EPA), and to determine the incidence and level of pesticide residues in the food supply. FDA uses two general food supply monitoring systems to address these objectives. These are general commodity monitoring, and the so-called "Total Diet Study." General commodity monitoring involves sampling on an "as shipped" basis raw agricultural commodities, processed foods, and animal feeds. These samples are analyzed for the purpose of enforcing tolerances established by EPA, and determining the incidence and levels of residues. The Total Diet Study involves collecting a "market basket" of food samples four times per year and analyzing the foods in a ready-to-eat form. The Total Diet Study is used to estimate dietary intake of selected pesticides by various U.S. age-sex groups.1 1 Reed, Donald, Pasquale Lombardo, John Wessel, Jerry A. Burke and Bernadette McMahon. The FDA Pesticide Monitoring Program (Unpublished draft, available through the Division of Chemical Technology, Food and Drug Administration). CRS-2 ISSUE AND BACKGROUND During the past 10 years, several reports have addressed issues concerning the analytical methods used in the FDA Pesticide Monitoring Program. In the most recent of these, the General Accounting Office (GAO), in 1986, reported that there was a significant gap between the number of pesticides that could potentially be found in food, and the number that the FDA is able to detect practicably with classical laboratory methods now in use.2,3 while the size of the gap and its significance from a public health standpoint is controversial, there is a consensus, in which FDA concurs, that more practicable methods are needed. The GAO studies reported several weaknesses in the general comodity monitoring system, some of which were linked to limitations of existing pesticide analytical methods. FDA comments on the reports reflected substantial agreement with the GAO observations.4 GAO reported that the monitoring program made use of two general types of pesticide analytical methods: multiresidue and single residue methods. 2 U.S. General Accounting Office. Pesticides: Need to Enhance FDA's Ability to Protect the Public From Illegal Residues; Report to Congressional Requesters by the Comptroller General of the United States. GAO/RCED-87-7. Washington, D.C., 1986. p. 31-33. [Hereinafter cited as, Domestic Food Report]. 3 U.S. General Accounting Office. Pesticides: Better Sampling and Enforcement Needed on Imported Food; Report to the Honorable Frank Horton by the Comptroller General of the United States. RCED-86-219, Washington, D.C., 1986. 56 p. [Hereinafter cited as, Imported Food Report]. 4 Bowen, Otis R. Secretary of Health and Human Services. Letter to Charles Bowsher, Comptroller General of the United States. Mar. 20, 1987. [Hereinafter cited as, Bowen Letter to GAO, Mar. 20, 1987.] CRS-3 Multiresidue methods are capable of detecting a large number of pesticides having similar chemical and physical properties in a test of a single sample. FDA has developed five broad scope multiresidue methods which are suitable for routine use in the PMP. Single residue methods are capable of identifying only one pesticide residue in a food sample. In general, multiresidue and single residue methods require comparable time and resources to conduct per sample. Therefore, multiresidue methods are considered more time and resource efficient than single residue methods. GAO reported that, because of resource constraints, food samples collected through the PMP are generally analyzed using only one of the five multiresidue methods developed by FDA. Single residue methods are reserved for use in conducting special surveys; confirming the presence of illegal residues detected by multiresidue methods; and testing for specific pesticides not detected by multiresidue methods, when FDA knows or suspects it is present in food. GAO reported that FDA would have to either significantly reduce the amount of sampling and testing, or dramatically increase resources if single- residue methods were to be used routinely.5 Each multiresidue method is capable of detecting only certain pesticide residues in selected food classes. The scope of coverage of the five commonly used methods ranges from 24 to 123 different residues. Together, the tests are capable of detecting 203 different pesticide residues. GAO reported that the most serious limitation it observed concerning the methods, was that cumula- tively the methods are capable of detecting only 40.9 percent of the 496 different pesticides that FDA has listed has having residue potential. These methods detect approximately 64 percent of 316 pesticides that have food 5 Domestic Food Report, p. 31-33. CRS-4 tolerances in force and are either currently registered for use on food, or persist in the environment and appear in food despite cancellation/suspension of food uses.6 Therefore, for up to 59.1 percent (293) of potential pesticide residues, single residue methods must be used.7 Because single residue methods are not used by FDA for routine monitoring, pesticides not detected by the five multiresidue methods are not routinely monitored. FDA has disputed the relevance of comparing the scope of coverage of the five multiresidue methods to the reference point GAO chose (496 pesticides). FDA contends that the number GAO relied upon counts pesticides that are not used because registration is pending or has been cancelled, and pesticides for which all food uses have been cancelled.8 The agency has implied that 230 may be a more appropriate reference point.9 Nonetheless, they agree that more multiresidue methods are needed.10 Furthermore, GAO reported that among those pesticides that FDA has classified through its Surveillance Index as warranting routine monitoring, 33 of 81 are not covered by any of FDA's five most commonly used multiresidue methods. They must be detected using less comprehensive or single residue methods. However, GAO reported that of these 33 pesticides, 30 6 Domestic Food Report, p. 31-34. 7 Ibid., p. 33. 8 Bowen Letter to GAO, Mar. 20, 1987. p. 1. 9 Ibid., p. 2. 10 FDA Staff of the Chemical Technology, Field Science, Federal State Relations Divisions and the Contaminants Policy Staff. (Mr. Jerry Burke, Pasquale Lombardo, Paul Corneliussen, Dr. Arvin Shroff, et al.) Private interview. lRockville, Md., Apr. 1, 1987. [Hereinafter cited as FDA Meeting, Apr. 1, 1987]. CRS*5 of these received little or no testing in domestic foods during the period 1979 through 1985.11 The Surveillance Index is a system developed by FDA beginning in 1979 to classify pesticides according to potential health hazard. The classification scheme was developed to help the agency prioritize its monitoring of residues not detected by multiresidue methodslz and to target research efforts to develop new methods.13 Pesticide chemicals undergoing "special review" at EPA because of special health or environmental concerns have been given priority in being classified in the Surveillance Index. The Index lists pesticides in Classes 1, II and III, with Class I being of highest priority, based upon the ‘toxicity, use pattern (poundage and crop), potential for food residues and persistance in the environment.14 FDA has classified 186 pesticides. Among those pesticides not covered by the multiresidue methods commonly used by FDA are 8 Class I pesticides. Class I pesticides are considered to pose a high health hazard warranting immediate inclusion in the pesticide monitoring program on a continuous basis.15 In addition, several Class II 11 Domestic Food Report, p. 37-38. 12 Reed, et al., The FDA Pesticide Monitoring Program, p. 12. 13 U.S. Dept. of Health and Human Services. Food and Drug Administra- tion. FDA Monitoring Programs for Pesticide and Industrial Chemical Residues in Food; Study Group on FDA Residue Programs, June 1979. p. 57. 14 Domestic Food Report, p. 35. 15 Reed, Donald. The FDA Surveillance Index for Pesticides: Establishing Food Monitoring Priorities Based on Potential Health Risk. Journal of the Association of Official Analytical Chemists, v. 68 (1982). p. 22. CRS*6 pesticides are not covered, including the widely usedlb ethylene bisdithiocarbamates (EBDC's) Mancozeb, Maneb, Metiram and Zineb.17 Pesticides classified as Class II are those FDA considers to pose possible high risk toxicity effects combined with the potential for significant human dietary exposure. The hazard is considered sufficient to warrant a temporary inclusion iof the pesticide in the monitoring program as soon as possible, and to continue until exposure is better defined or additional toxicity data are available.18 16 USDA has estimated that EBDCs account for approximately 22 percent of the total fungicides used in the United States and 57 percent of those used in the world. Donald Reed, Food and Drug Administration (internal document) Dec. 19, 1979. Mancozeb is used on approximately 80 percent of the onion acreage in the United States and on apples, potatoes and tomatoes. Environmental Protection Agency Pesticide Fact Sheet: Mancozeb, Apr. 1987. 17 Several EBDCs have been demonstrated to induce lung and thyroid tumors- in experimental animals. Donald Reed, Food and Drug Administration (internal document) Dec. 19, 1979. 18 Reed, The FDA Surveillance Index for Pesticides. CRS-7 HISTORY OF RECOMMENDATIONS CONCERNING DEVELOPMENT OF ANALYTICAL METHODS The GAO reports of 1986 were not the first time that the limitations of existing pesticide analytical methods were brought to the attention of Congress. Significant investigations of the Pesticide Monitoring Program were made by the Subcommittee on Oversight and Investigations of the House Committee on Interstate and Foreign Commerce and CAD during the late 1970s. These prompted FDA to conduct its own critical review of the program. The FDA study generated recomendations which have become the touchstone for program improvements. In 1978, the subcommittee recommended that FDA develop analytical methods to detect more substances, and methods that could be performed more quickly. In addition, the subcommittee recommended that FDA use all detection methods available to it, including single residue testing where multiresidue methods could not be used.19 In the 1979 report of the FDA Study Group, the recommendations emphasized the importance of a "strong, continuously well-supported and closely coordi- nated analytical methods development program" to the effectiveness of the Pesticide Monitoring Program.20 The Study Group stated that practical methods 19 U.S. Congress. House. Committee on Interstate and Foreign Commerce, with separate views by the Subcommittee on Oversight and Investigations. Cancer-Causing Chemicals in Food. 95th Cong. 2d Sess. Washington, U.S. Govt. Print. Off., 1978. p. 37-38. Committee Print. 20 U.S. Dept. of Health and Human Services. Food and Drug Administration. Public Health Service. Study Group on FDA Residue Programs.. FDA Monitoring Programs for Pesticide and Industrial Chemical Residues in Food. June 1979. p. 57. [Hereinafter cited as 1979 FDA Study Group Report.] CRS-8 were needed so that FDA could analyze sufficient numbers of samples. It also recomended that FDA continue to develop new and expanded multiresidue methods, and that resources also be devoted to conducting research on other analytical approaches: rapid single residue analytical methods, bioassay screening and automated analysis.21 The Study Group recommended that FDA prioritize the needs for new methods based upon the Surveillance Index classification of pesticides. The Surveillance Index scheme was first proposed in the 1979 report. The recommendations of the 1979 report were intended to reflect the "ideal" program for methods development without regard to the resources that implementation would entail.22 In 1981, FDA again addressed the need for new analytical methods in its response to program changes recommended by the Senate Appropriations Committee in its report on the FY82 FDA appropriations. FDA reported that it had reprogrammed resources in 1980 and 1981 from other food safety activities to augment methods development in the chemical residue area. However, FDA explained that further reprogramming was unlikely to occur.23 Inga 1983 Status Report on the Study Group recommendations (the most recent report), FDA reported that the Surveillance Index had been established and was bein used to tar et develo ment of more "classical" anal tical methods 8 8 P Y 21 1979 FDA Study Group Report, p. 58-59. 22 FDA Study Group Report, preamble. 23 U.S. Department of Health and Human Services. Food and Drug Administration. Public Health Service.‘ Report on Proposals to Improve Control of Pesticide Residues in Imported Foods. Mar. 10, 1982. p. 9. CRSf9 (see p. 10-1l).24 However, FDA rescinded25 the recommendation that resources be devoted to other analytical approaches, primarily because of "budget uncertainties" which made it "[im]possible to assign resources specifically for basic research on other analytical approaches."26 24 U.S. Dept. of Health and Human Services. Food and Drug Administra- tion. Public Health Service. Steering Committee on FDA Residue Programs. FY83 Status Report on Recommendations of Study Group on FDA Pesticide and Industrial Chemical Residue Programs. Rockville, Md. Mar. 8, 1983. [Hereinafter cited as 1983 Residue Program Status Report]. 25 "Rescinded" was defined in the report to mean "[t]he recommendation has been reevaluated and is no longer considered necessary." 1983 Residue Program Status Report, p. ii. 26 1983 Residue Program Status Report, p. 11. CRS-10 FDA'S CURRENT RESEARCH PRIORITIES IN ANALYTICAL METHODS DEVELOPMENT FDA describes its current analytical methods development work as "applied research and highly mission oriented."27 Most work is directed toward immediate program needs within the framework of five general goals: 1. expanding the coverage of existing multiresidue analytical methods to additional pesticides and alteration products; 2. extending methods to cover different food/feed comodities; 3. validating analytical methods; 4. adapting and integrating new analytical methods into existing ones; and 5. developing "new" analytical methods or techniques.28 FDA's FY87 research Technical Plans reveal how the above listed goals are currently being pursued. FDA is currently developing multiresidue methodolo- gies for two types of organohalogen pesticides: pyrethroids and phosphines and metal phosphides.29 In addition, FDA is assessing the feasibility of modifying current methods for organophosphorus pesticide analysis by revising sample preparation ("clean-up").30 A confirmation test (derivatization) for an 27 Corneliussen, Paul E. Development of Analytical Methods for Pesticide Residues. Food and Drug Administration, Pesticide and Industrial Chemicals Branch. Apr. 1, 1987 (internal document). E 23 Ibid. 29 Technical Plan FY87, Chemical Contaminants. Organohalogen Pesticides, project code 00180. 30 Technical Plan FY87, Chemical Contaminants. Organophosphorus Pesticides, project code 00191. CRS-ll official analytical method for a breakdown product (Ethylene thiourea, "ETU") of the organonitrogen fungicides, EBDC's (see page 6), is also under study.31 A method is being developed for triphenyltin32 and its breakdown products.33 All of the above projects make use of "classical" analytic techniques. FDA's work on "new" (e.g., bioassay) approaches is limited in FY87 to "monitoring the status of immunoassay applications to analysis of food for pesticide and industrial chemical residues.34 FDA recently reported that it is now establishing a contract for the development of a few (unspecified) immunoassays for pesticides in food. FDA envisions funding this project at $200,000 for the first year with additional funding probable in subsequent years.35 31 Technical Plan FY87, Chemical Contaminants. Organonitrogen Pesticides, project code 12668. 32 Triphenyltin is considered a very toxic chemical. Triphenyl Acetate and Hydroxide are used as agricultural fungicide, particularly in Europe. Several human poisonings from occupational exposures have been reported. Symptoms included headache, nausea, vomiting, epigastric pain, high blood sugar, transient loss of consciousness and possible convulsion. Gosselin, Robert E., Roger P. Smith and Harold C. Hodge. Clinical Toxicology of Commercial Products. Baltimore, Williams and Wilkins, 1984. p. I-2, II-148. 33 Technical Plan FY87, Chemical Contaminants. Methodology for Pesticides of High Interest, project code 10311. (Triphenyltin. . . . 34 Technical Plan FY87, Chemical Contaminants. Analytical Capabilities of FDA Pesticide Monitoring, project code 12851. Quarterly Project Progress Report (lst Qtr./FY87), Technical Plan FY87; project code 12851. 35 Written communication. Charles Banks. Legislative Affairs Office, Food and Drug Administration, Apr. 22, 1987 (443-3793). cas-12 CURRENT ISSUES IN METHODS DEVELOPMENT The 1986 CAO reports together with the earlier congressional, GAO and FDA studies of the 1970s, found a longstanding and persisting deficiency in the scope of coverage of analytical methods practicable for use in FDA's Pesticide Monitoring Program. Some progress has been made in complying with recommended program changes, (e.g., the establishment of the Surveillance Index). However, in other areas, limited progress has been reported. In 1977, CAO reported to the Subcommittee on Oversight and Investigations that FDA's five most commonly used multiresidue tests covered 107 of 268 (approximately 40 percent) of pesticide chemicals with established food to1erances.36 In 1986, GAO reported that the five multiresidue methods covered 203 of the 496 (40.9 percent) of pesticide chemicals with residue potential, and approximately 64 percent of pesticides with permanent food tolerances (p. 3-4). In addition, FDA has not ' and as noted developed either single or multiresidue "rapid test methods,‘ above, rescinded the recommendation that research be conducted in this area. The persistence of the gap between pesticides in use and those that can be practicably detected has caused observers to question: why are there not more multiresidue methods, and what, if any, are the applications of rapid test methods? 36 Committee report on Cancer-Causing Chemicals in Food, p. 33. CRS-13 MULTIRESIDUE METHODS what are Multiresidue Methods? As noted above, multiresidue methods are analytical tests capable of detecting more than one residue in a single sample. ‘The primary multiresidue methods used in FDA's Pesticide Monitoring Program are of broad scope. Each method detects many pesticides in one or more classes and each is adaptable to different foods within a food class (e.g., high fat foods). These "classical" methods are listed below37: 1. Luke Method38 Foods: Fruits/Vegetables/Some Cereal Grains (limited to foods in these classes that require minimal "clean-up," e.g., foods without excessive pigmentation). Pesticides: Covers ionic and nonionic pesticides.39 Number of Pesticides Coveredzu 121 2. Fatty Food Method4Q Foods: Dairy/Meat/Fish. Pesticides: Nonionic pesticides in the organophosphorus and organochlorine classes. Number of Pesticides Covered: 97 37 Corneliussen, Paul. Personal Communication. Pesticides and Industrial Chemicals Branch, Division of Contaminants Chemistry, Food and Drug Administration. Apr. 17, 1987. (245-1231) 38 U.S. Dept. of Health and Human Services. Food and Drug Administration. Pesticides Analytical Manual Volume I: Methods which Detect Multiple Residues. Section 232.4. [Hereinafter cited as PAM I]. 39 The terms "ionic" and "nonionic" refer to a chemical property of a pesticide which influences whether it will be soluble in water or fat respectively. 40 PAM 1, section 211.1. CRS-14 3. Nonfatty Food Method41 Foods: Fruits/Vegetables (suitable for food samples requiring more clean- up than the Luke method, e.g., heavily pigmented foods). Pesticides:. Nonionic pesticides in the organophosphorus and organochlorine classes. Number of Pesticides Covered: 123 4. Storhrerr Method42 Foods: Fruits/Vegetables Pesticides: Nonionic pesticides in the organophosphorus class. Number of Pesticides Covered: 61 y 5. Krause Method43 Foods: Fruits/Vegetables Pesticides: Ionic pesticides in the organonitrogen class (N-methylcarbamates) Number of Pesticides Covered: 24 The term "method," as used above, refers to the sequence of the following general analytic techniques: Extraction. A food sample is prepared, e.g., homogenized and chemically processed, so that fraction (e.g., fat component) that is likely to contain pesticide chemicals is removed. Clean-up. The extracted fraction is processed to remove material (e.g., food pigments) that may interfere with pesticide identification and quantification. The extraction and clean-up phases are particularized for analysis of different categories of food and classes of pesticides. Therefore, multiresidue methods are named (e.g., "fatty method") for their distinguishing extraction and clean-up steps.44 41 PAM I, section 212.1. 42 PAM 1, section 232.3. 43 PAM 1, section 242.2. 44 Cornelliussen, Paul. Personal Communication, FDA. Apr. 17, 1987. CRS-15 Identification and Quantification of Residues. The technique used to analyze the residue contained in food extracts, for all but the Krause Method is gas-liquid chromatography (GLC). High-Pressure Liquid Chromatography (HPLC) is used in the Krause method. Both GLC and HPLC operate on a similar basic principle. A chemical mixture (food sample extract) is injected into the portal of specialized chromato- graphic equipment which contains a "separating column." The separating column permits pesticides to pass through it at different rates based on differences in their physical and chemical properties. As each chemical reaches the end of the column, each is in turn sensed by a detector and a printer displays the chemical as a "peak" on a chromatographic print-out. Chemicals can be identified by comparing the time interval between sample injection and the appearance of a peak, with the intervals of standard known chemicals. The amount of a chemical is determined by measuring the size of the peak. Confirmation. After a residue has been identified using GLC (or HPLC for Krause Method), the identities of the residues must be confirmed. The test methods used for confirmation vary with the tentative identity of the pesticide, the amount of residue available for testing, the sample type, and availability of equipment required for tests. A technique known as "mass spectrometry" is often used as a confirmatory test for GLC. Detailed guidelines for choosing the appropriate confirmatory tests are set out in the Pesticide Analytical Manual, Volume I.45 45 PAM 1, section 601. CRS-16 Why Are There Not More Multiresidue Methods? There appear to be two primary impediments to developing additional multiresidue methods practicable for use in FDA's monitoring program: science and resources. The scientific challenge of developing new methods has grown during the past 15 years. The primary multiresidue methods were developed at a time when most of the pesticides of concern were of the organophosphorus and organochlorine classes (e.g., DDT, chlordane and malathion). This fact made it possible to develop multiresidue methods of broad scope while targeting the method for only one or two chemical classes. However, because of health and environmental hazards associated with pesticides in these classes, many have been phased out of use, and newer pesticides are of more diverse chemical classes. In addition, many new pesticides (considered safer, in part because thay are less persistent in the environment) degrade more quickly into "alteration" by-products. This means that to adequately monitor exposure to a single pesticide, methods may be needed to detect several alteration products. Together, these changes in pesticide formulation have increased the number and chemical class diversity of compounds to be analyzed, significantly increasing the scientific task of developing adequate numbers of multiresidue methods.46,47,48 A common perception among residue chemists is that it may no longer be possible to develop multiresidue methods capable of detecting 50 or 100 compounds. Multiresidue methods for food that are currently being developed 46 FDA Meeting Apr. 1, 1987. 47 Trichilo, Charles and Francis Griffith. Residue Chemistry Branch. Office of Pesticide Program, Environmental Protection Agency. Private Interview. Crystal City, Va., Apr. 2, 1987. [Hereinafter cited as EPA meeting, Apr. 12, 1987.] 48 Personal Communication, Dr. David MacLean, Association of Official Analytical Chemists, Mar. 30, 1987 (522-3032). CRS-l7 are likely to detect only a handful of compounds. For example, the multiresidue method FDA is developing for pyrethroids (see p. 10) will probably be capable of detecting 6 or 7 compounds.49 Although additional pyrethroid pesticides will probably be developed in the future, and be detected by the method, it is likely to remain of relatively narrow scope. The second major challenge to developing new methods is cost. It generally takes between 8 to 10 years to develop a new multiresidue method.50 It also requires highly trained residue chemists. Because of the resource investment needed to develop these methods, Statessl and private industry52 generally are not involved in such work. Among Federal agencies having responsibility related to pesticides, FDA is considered the "lead agency" for developing new methods for most food products. Research Activities Food and Drug Administration FDA's current research activities are described at p. 8. [The FY87 budget estimate for FDA's methods development program has not yet been provided by the agency.] 49 Personal Communication. George Yip, Pesticide and Industrial Chemicals Branch, Division of Contaminants Chemistry, Food and Drug Administration, Apr. 20, 1987. (755-1649). 50 FDA Meeting, Apr. 1, 1987; EPA Meeting, Apr. 2, 1987. 51 Dr. Francis Griffith, Residue Chemistry Branch, Office of Pesticide Program, Environmental Protection Agency. Private Interview, Apr. 2, 1987, Crystal City, Va. (557-7342). 52 Jack Cooper and Edward Elkins, Environmental Affairs Division, National Food Processors Association, Private Interview, Apr. 3, 1987, Washington D.C. (639-5925). [Hereinafter cited as National Food Processors Meeting, Apr. 3, 1987.] CRS-18 U.S. Department of Agriculture USDA's research on multiresidue methods for pesticide residues in food is limited. The Food Safety and Inspection Service (FSIS) of USDA is responsible for monitoring pesticide residues in meat, poultry, and egg products.53 FSIS funds, by contract, research designed to adapt multiresidue methods to the special characteristics of meat and poultry samples. In addition, USDA's "in- house" system of laboratories, the Agricultural Research Service (ARS), does some multiresidue development to meet FSIS priorities under a memorandum of understanding with FSIS (funding level currently unavailab1e).54 The funding level of FSIS contracts for development of multiresidue methods has fluctuated during the past five years. In FY82, FSIS spent $590,536 on contracts. No funding was provided in FY83. In FY84, $201,806 was spent, and it FY85 and FY86, $201,806 and $275,767 were spent respectively.55 The Cooperative States Research Service (CSRS) of USDA, which funds agricultural research through universities and organizations, has devoted no funds to the development of multiresidue methods.56 53 Meat Inspection Act (21 U.S.C. 601 gt. gg3.); Poultry Products Inspection Act (21 U.S.C. 451 gt. g_g.); Egg Products Inspection Act (21 U.S.C. 1031 gt. g_g.). 54 Hill, Kenneth. Personal Comunication. Agricultural Research Service, U.S. Dept. of Agriculture, Mar. 31, 1987. (344-2495). 55 Information provided by: Mr. Bill Nest, Office of Budget and Finance, Food Safety and Inspection Service, U.S. Dept. of Agriculture, Apr. 17, 1987. (447-3367). 56 Guilliland, Betty. Personal Communication. Budget Office, Coopera- tive States Research Service, U.S. Dept. of Agriculture, Apr. 14, 1987, 447-5787. CRS-19 The Environmental Protection Agency Since 1984, three proposals have been made within EPA to develop multiresidue methods for foods. The proposed methods were aimed at triazine herbicides (e.g., atrazine), alachlor herbicides and synthetic pyrethroids. None of these proposals were funded, however.57 The Residue Chemistry Branch of the Office of Pesticide Program (OPP) validates58 single residue methods for foods that are submitted by pesticide registrants. Companies seeking registration (approval) of a pesticide must submit a residue method for enforcement of tolerances whenever a food tolerance is proposed for the pesticide.59 These methods are "tried-out" by EPA to determine whether they work. The methods are then published in FDA's Pesticide Analytical Manual, Volume II (PAM II). Registrants constitute the greatest source of new single residue methods. However, as discussed above, such methods have only limited application in FDA's Pesticide Monitoring Program. In addition, "PAM II" methods often have been criticized as being impractical, costly and unreliable to use. Critics observe that methods sometimes require exotic equipment or reagents that are 57 Griffith, Francis. Personal Communication. Residue Chemistry Branch,Office of Pesticide Program, Environmental Protection Agency, Apr. 23, 1987. (557-7324). 58 "Validation" here refers to a method "try-out" conducted by the EPA laboratory. This process should be distinguished from the "validation" required by the Association of Official Analytical Chemists (AOAC) for a method to gain "official" status (the "gold standard" of analytical methods). AOAC requires that at least six laboratories validate a method. The purpose of validation by AOAC criteria is to assure that the method is capable of performing as intended and that the results of an analysis are of acceptable accuracy and precision. See, Hill, Kenneth R. and Paul E. Corneliussen, Validation of Official Methods, Analytical Methods for Pesticides and Plant Growth Regulators, v. XV, 1986. p. 111-132. 59 40 C.F.R. 158.125. CRS-20 not readily available, and that the EPA validation process is not rigorous enough to assure the method will work.6O In response to these criticisms, EPA has begun to evaluate ways to improve the validation of methods.61 Registrants Must Evaluate Multiresidue Methods In 1984, EPA promulgated a regulation that requires pesticide registrants to submit data on whether the FDA/USDA multiresidue methods detect and identify the pesticide for which a tolerance is sought.62 The purpose of this regulation is two-fold. First, the experiments performed by registrants may show that the pesticide to be registered can be analyzed using existing multiresidue methods. Second, even if the methods cannot accurately quantify the pesticide residue, one or more may be capable of detecting the pesticide so that a peak is displayed on the chromatographic print-out. In such a case, the peak size would be an unreliable indicator of the amount of residue. However, the location of the peak (i.e., reflecting the interval between sample injection and peak display) could be used to identify the residue. This information is expected to be helpful to residue chemists testing samples of unknown history. It should help chemists to identify "stray peaks" that appear on the chromatograph, and to help them choose the appropriate single residue methods to be used to quantify the pesticides producing stray peaks. 60 Cooper, Jack and Edward Elkins, Environmental Affairs Division, National Food Processors Association. Private Interview, Apr. 3, 1987 (639-5925). 61 Griffith, Francis. Draft Issue Paper: Options for Analytical Enforcement Method Tryout. (Memorandum to John Melone, Director, Hazard Evaluation Division, Office of Pesticide Program, Environmental Protection Agency) Sept. 9, 1986. 62 40 CFR 158.125. CRS-21 The new regulation has not yet been fully implemented63, so it is too early to evaluate its impact. However, the regulation is expected to enlarge the scope of existing multiresidue methods, and to help make appropriate single residue methods easier to select and, thus, more practical to incorporate into the Pesticide Monitoring Program.64 Limitations of Multiresidue Methods Multiresidue methods are highly sensitive and are the current methods of choice for residue analysis. However, these methods have several limitations. The sensitive nature of chromatographic equipment requires that extensive sample preparation (extraction and clean-up) be done before pesticides can be identified and quantified. The techniques require specialized equipment and controlled laboratory conditions. The chromatographic equipment can analyze different samples only sequentially, not simultaneously, so that the amount of equipment available becomes a limiting factor in the rate of analysis. Also, the techniques can be carried out accurately only by highly skilled analytical chemists with specialized experience in residue chemistry.65 These factors cause traditional residue analysis to be costly. FDA estimates that analysis of fruit, vegetable and grain samples costs up to $200 per sample. Fatty foods, including milk and fatty fish cost approximately $320 per sample. Some 63 Pesticide Assessment Guidelines Subdivision O-Addendum; Availability of Final Guidance Document for Analytical Methods for Multiresidue Protocols, 51 FR 34249, Sept. 26, 1986. 64 EPA Meeting, Apr. 2, 1987. 65 Griffith, Francis. Private Interview. Residue Chemistry Branch, Office of Pesticide Programs, Environmental Protection Agency. Crystal City, Va., Apr. 2, 1987 (557-7324). CRS-22 chemicals are especially difficult to analyze. Probably the most expensive analysis is determining dioxin in a fish sample, which costs about $1,700 and takes about 3 weeks to run.66 RAPID TEST METHODS The limitations of traditional analytical methods have stirred interest in the potential application of "new" rapid test methods involving biologically produced reagents.67 As noted above, the 1979 report of the FDA Study Group (see p. 7-8 above) recommended that FDA explore possible applications of bioassays to pesticide residue analysis.68 Although FDA has not funded such research, it has been carried out by university scientists and private industry. The development of rapid test methods for pesticide residue analysis has followed two general paths: the development of enzyme tests and antibody tests. 66 Shroff, Arvin. Personal Comunication. Food and Drug Administration, Regulatory Affairs Division. Apr. 16, 1987 (443-6230). 67 A "reagent" is a substance used for the detection or determination of another substance. 63 The FDA Study Group also recommended that FDA explore more "rapid" techniques involving classical analytical methods (e.g., GLC), such as "automated analysis." Many techniques have been developed that can save time in residue extraction and analysis, e.g., robotic sample preparation, "tandem mass spectrometry," "universal extraction," etc. Such methods would speed the rate of analysis in the laboratory, but would not necessarily expand the number of pesticides covered by the methods. These techniques are not addressed in this report. CRS-23 Enzyme Inhibition Methods At least two companies69,70 have developed rapid screening tests designed to detect the residues of pesticides that inhibit the enzyme71, cholinesterase (ChE bioassay). Pesticides in the organophosphate and organocarbamate classes are effective against insect pests by interfering with the normal action of cholinesterase which is naturally present in the insect body. Cholinesterase is also present in the human body, and pesticide inhibition of the enzyme is responsible for certain human symptoms of pesticide exposure. The normal activity of cholinesterase is to cleave apart certain body chemicals (substrates) through a process known as "hydrolysis". The ChE bioassay works on the following principle. A substrate is selected that will produce a measurable effect when exposed to cholinesterase. For example, one test kit uses a substrate that turns blue when cleaved by cholinesterase. The substrate is exposed to a mixture of the enzyme and a fluid test sample (for fruits and vegetables, the sample is a surface rinse of the food). If a cholinesterase inhibiting pesticide is present in the sample, it will interfere with the action of the enzyme on the substrate. In the above example, it would prevent the substrate from turning blue. A blue test result would indicate a negative response.72 69 Smith, Ivan. Personal Communication. Enzytec Inc. (subsidiary of Midwest Research Inc., "MRI"), Apr. 13, 1987. (913)541-8585. 70 Ferguson, Bruce. Personal Communication. Immunosystems, Inc., Apr. 7, 1987. (207)282-4146. ‘ 71 An enzyme is a protein that triggers specific chemical changes. 72 Enzytech Pesticide Detection Systems Technical Bulletins. Product Bulletin, Pesticide Detector Ticket. 3/87, (available from Dr. Ivan Smith, (913) 541-8585); see also, Immunosystems Inc., Res-I-Dase, Instructions for Use (available from Dr. Bruce Ferguson, (207)282-4146). CRS-24 The advantages of the ChE bioassay are that the test is relatively inexpensive, quick, and easy to do. Each test costs between $4.00 and $7.00, and takes approximately five to 10 minutes to perform. Minimal training and equipment are required, which makes it possible for relatively unskilled people to use the test in the field.73 The reported limit of residue detection varies with the pesticide. One test kit vendor has reported detection of carbofuran (Furadan) at 0.1 ppm (part per million) weight/volume of sample; 0.2 ppm for aldicarb (Temik) and 20 ppm for metasystox-R. Several major food companies are reportedly satisfied enough with the test to use it regularly.74 However, the tests have not been offi- cially validated, so the reliability of the reported detection limit is uncertain. The limitations of the test are that it provides only a cumulative "reading" of whether any cholinesterase inhibitors are present. There are many such pesticides, and the test does not identify individual chemicals or quantify them. In addition, the method does not detect alteration products of the parent compound or the many pesticides that do not inhibit cholinesterase. The coverage of the bioassay overlaps with that of some of FDA's primary multiresidue methods. Unlike classical methods, which analyze a sample homogenate, ChE bioassay is capable of analyzing only a surface rinse of the food product. The value of surface residue monitoring is controversial. .Test vendors argue that surface residues, for most pesticides, are a good indicator of the pesticide "load" of a food sample. On the other hand, FDA argues that surface residues are 73 Ibid. 74 Smith, Ivan. Personal Comunication. CRS-25 inadequate, particularly for "systemic" pesticides (those taken up by the root systems of plants). The ChE bioassay is being promoted as a screening test. Vendors urge that FDA could afford to test a larger percentage of the imported and domestically produced foods if the bioassay were incorporated into the pesticide monitoring program. They argue that ChE bioassay detects the most commonly violative pesticide residues. Further, some argue that since most (they estimate at least 80 percent) of food samples are found not to be in violation of the tolerance when tested by classical methods, ChE bioassay offers a less expensive way to obtain these "negative results." Those samples found to be positive with ChE bioassay could be further analyzed using a classical method to identify and quantify particular residues, vendors contend. The bioassay is promoted also for identifying "residue free" foods in investigations of food contamination by cholinesterase inhibiting pesticides.75 Although the ChE bioassay is sometimes promoted as a cost-saving way to monitor for pesticides, it is not certain that it would save money if used in FDA's program. FDA argues that they cannot bring enforcement actions based on the test since it does not identify the violative pesticide(s), nor quantify them. Therefore, FDA has stated it would have to follow-up positive results with classical analytical methods.76 Furthermore, FDA does not currently monitor by way of "screening tests." To incorporate a screening test into the program would be to expand FDA's activities. In fact, if the food supply were screened, it is possible that more violative samples would be found, generating more samples to be tested with classical methods. On the other hand, the test 75 Ibid. 76 FDA Meeting, Apr. 1, 1987. CRS-26 may save money in identifying samples "negative" in surface residues in both routine monitoring activities and in investigating contamination episodes. If ChE be can be validated, and surface residue testing is determined to be meaningful, the potential contribution of the test would be to expand the number of samples that can be tested for cholinesterase-inhibiting pesticides for a given cost. The test appears to potentially offer a cost effective way to obtain negative results, but does not guarantee that, overall, dollars would be saved. Immunoassay Methods The second major type of rapid test method is immunoassay. Immunoassays have been used for several years to diagnose human conditions. Probably the most familiar of these are home pregnancy tests. However, only recently have tests been developed for detecting pesticide residues. The primary biological reagents used in an immunoassay are antibodies. Antibodies are proteins produced by the immune systems of humans and animals to fight off the intrusion of foreign matter (antigens), such as disease-causing organisms. One way antibodies participate in this "immune response" is to bind (attach to) to the antigen in a highly specific manner. Special antibodies are produced to match the unique characteristics of the antigen. Immunoassays in a sense capture this aspect of an "immune response" in a test kit. To produce antibodies to be used in a pesticide imunoassay, the pesticide of interest must be injected into the bloodstream of a laboratory animal. Mice and rabbits are used for this purpose. Pesticides are generally small molecules and cannot alone trigger antibody production. Antibodies are produced in response to larger molecules. The pesticide must be linked to a CRS-27 large molecule (usually a protein) which together can elicit antibody production specific to the free pesticide.77 After the immune response is obtained, antibody-producing cells are removed from the spleen of the laboratory animal. The cells are screened to select the cells that produce antibodies that bind most specifically to the pesticide chemical of interest. Each cell is specialized to produce only one antibody type. For "monoclonal antibody" tests, only the cells that produce one antibody type are selected. For "polyclonal antibody" tests, cells that produce different antibody types (all in response to the pesticide of interest) are selected. These spleen cells are fused with rapidly dividing cells (myeloma) becoming a hybridized cell known as a "hybridoma." The hybrid cells grow rapidly in culture, which allows the production of large quantities of the antibody(ies) of interest. The antibodies can be extracted from the cells for use in immunoassays. Some of the antibody-producing cells are reserved and frozen. The reserved cells can be cultured at a later time and serve as a constant supply of the antibodies of interest.78 All immunoassays operate according to the same fundamental principle. A known quantity of the antibody specific for a particular pesticide is exposed to a test sample. If the pesticide is present, the antibody is highly sensitive to it and will bind to it. A test result is quantified by either measuring the amount of antibody bound to the pesticide, or the amount that remains free. There are several ways the antibody can be measured. One technique is to link the antibody binding reaction to a color change. For example, a color change would occur when pesticide is present. The density of 77 Mccilvery, Robert W. and Gerals W. Goldtein. Biochemistry, A Functional Approach. Philadelphia, W.B Saunders Co., 1983. p. 226-231. 73 Ibid. CRS-28 the color will correspond to the concentration of the pesticide. This information can potentially reveal not only whether the pesticide is present, but give the quantity present.79 Some investigators are exploring ways to link immunoassays with computer technology so that results can be directly displayed as a numerical reading. This technique is sometimes referred to as a "biosensor." if perfected it would make it easier to quantify pesticide residues using immunoassay. Most tests that have been developed so far are "monoclonal antibody tests," but for certain projects, "polyclonal" tests are more sensitive.80 Antibodies can be produced to detect single pesticides, or a common characteristic of a class of pesticides. The strengths and limitations of immunoassays are closely intertwined. Some of the limitations of current tests will probably be overcome in the future. Immunoassays are highly sensitive, Specific, rapid_and inexpensive. This means that the tests can rapidly detect, in a relatively crude mixture, very small amounts of a specific pesticide. Immunoassays may be used on less refined samples than classical methods, which may save time and money in sample clean-up and extraction. They can also be run in the field on rinses of surface residues on food (this again raises the question of the value of surface residue data). Once a sample is prepared, it normally takes less than 10 minutes to run the assay. 79 Vanderlaan, Martin. Immunoassays for Trace Organic Analysis and Human Health Monitoring. In: Peer Review and Biomarkers Panel Meeting, Meeting Summary Report. Sept. 30, 1986. (Available through the Environmental Protection Agency). 30 Hammock, B. D., S. J. Gee, P. Y. K Cheung, et al. Utility of Immunoassay in Pesticide Trace Analysis. (Available from Dr. Bruce Hammock, University of California, Davis, Calif. 95616). CRS-29 Immunoassay tests are also adaptable to almost any pesticide. Antibodies can be produced for any pesticide, except for those of extraordinarily small molecule size (the trend is toward developing pesticides with a larger sized molecule). However, immunoassays are most effective in testing residues that are more soluble in water. This characteristic appears to make immunoassay a complementary technology to classical methods, which are generally most effective on more fat soluble compounds. This principle appears to operate to make immunoassay most cost effective for pesticides that are difficult, and \ therefore more expensive, to analyze by classical methods. For example, one laboratory reported that paraquat analysis using a classical method (GLC) cost approximately $500 per sample, while the same analysis using an immunoassay cost 35 cents per sample.81 A commercially available immunoassay designed to screen for organochlorines or atrazines costs about $12 to $15 dollars per test.82 Immunoassays can be designed to meet a variety of needs. They can be used as screening tests to detect the presence of a residue of a specific pesticide or of a class of pesticides (by detecting a characteristic common to the pesticide class). However, a class-specific immunoassay cannot identify or quantify individual pesticides. Some of the primary attributes of immunoassay, i.e., sensitivity and specificity, are limitations in situations where one wishes to test for several pesticides at once. Immunoassays are difficult to transform into multiresidue methods. Some investigators have tried to combine antibodies to several 81 Hammock, Bruce. Personal Communication. Professor of Entomology and Environmental Toxicology, University of California, Davis, April 16, 1987. (916)752-7519. 82 Immunosystems Inc., Res-I—Mune, Instructions for Use (available from Dr. Bruce Ferguson (207)282-4146). CRS-30 different chemicals to create a "multiresidue" method. However, such an approach produces interference between the antibodies which can destroy the reliability of the test if too many different antibodies are used. In appears that antibody mixtures may be useful as multiresidue methods as long as one limits the test to approximately 6 to 12 compounds.33 Given that classical multiresidue methods currently under development are likely to have about the same scope (e.g. method for pyrethroids), immunoassays may have at least limited application as multiresidue methods. An additional adaption of immunoassay to detect several residues is to design the test kit so that several individual antibody tests can be run simultaneously.84 Another limitation of immunoassays is that they are sometimes prone to "cross-react" with pesticides that are similar chemically. Some investigators believe that this does not pose a significant difficulty. They point out that cross-reactivity has been successfully controlled in human diagnostic iapplications of immunoassay.85 In addition, one investigator has capitalized on cross-reactivity among certain pesticides in developing a technique which may have potential as a multiresidue method.86 The extensive use of immunoassay in human medicine has provided a body of experience which provides the foundation for development of environmental applications. While pesticide immunoassay investigators maintain that the method is not a panacea, they point to several areas where it may be used 33 Personal Communication. Dr. Llewellyn Williams, Environmental Monitoring Systems Laboratory, Environmental Protection Agency, Apr. 16, 1987. (702)798-2138. 34 Ibid. 35 Hammock, Bruce. Personal Communication. 86, Williams, Llewellyn. Personal Communication. CRS-31 beneficially. Among those include: analysis of compounds difficult to analyze by classical methods (e.g., more water soluble compounds), analysis of compounds not detected by classical multiresidue methods, rapid analysis of large numbers of samples, field analysis,.and analysis of new large molecule pesticides produced by genetic engineering (e.g., proteins) which are not analyzed using classical techniques.87 Research Activities Federal and State agencies as well as private industry have expressed interest in immunoassay, but many are uncertain about how it ultimately may be applied. These organizations also have been only minimally involved in doing research on the application of immunoassay to pesticide analysis. Most work of this type has been done in academic laboratories and a handful of firms. Firms attempting to enter the market complain that enforcement agencies such as the FDA have been too slow to appreciate the benefits the method offers. In addition, they urge that a showing of Federal interest is needed to speed development of appropriate immunoassay applications. FDA, in particular, appears skeptical that immunoassay can be adapted to the diverse food samples they must test. This fact, together with resource constraints, may explain why FDA currently has no research effort on immunoassay (see ps 9). No pesticide immunoassays have been officially validated (i.e., using AOAC procedures). Representatives of Federal agencies88 the food industry,89 and 87 Hammock, Utility of Immunoassay in Pesticide Trace Analysis. 88 FDA meeting, Apr. 1, 1987. 89 National Food Processors Meeting, Apr. 3, 1987. CRS-32 academic communitygo have emphasized the need for validation of the methods before they can be relied on. Federal Activities in Rapid Pesticide Analysis As noted above, FDA's activities concerning rapid test methods are limited to monitoring the development of immunoassays for pesticide residue analysis (see p. 11). USDA estimates that for FY87, the commitment to immunoassay research (including assays for drugs, microbiological agents, and pesticides) is $40,000 through the Cooperative States Research Service91 and approximately $1.6 million through the Agricultural Research Service.92 FY87 figures are currently unavailable for the Food Safety and Inspection Service. However, a research plan is being developed with Lawrence Livermore Laboratory to develop immunoassay methods for chlordane and heptachlor.93 In addition, in FY86 $83,000 was spent on developing a rapid screening test for 14 organohalide pesticides and for polychlorinated biphenyls (PCBs).94 EPA has not funded research on rapid test methods for pesticide analysis in foods. However, the EPA Environmental Monitoring Systems Laboratory has 90 Hammock, Bruce. Personal Communication. 91 Cuilliland, Betty. Personal Communication. U.S. Dept. of Agriculture, Cooperative States Research Service, Budget Office. Apr. 14, 1987 (447-5787). 92 Cole, Darrel.i Personal Communication. U.S. Dept. of Agriculture, Agricultural Research Service. Apr. 20, 1987 (344-3861). 93 Hoffman, Michael. Personal Communication. U.S. Dept. of Agriculture, Food Safety and Inspection Service, Apr. 21, 1987 (447-7623). 94 West, Bill. Personal Communication. ‘U.S. Dept. of Agriculture, Office of Budget and Finance, Food Safety and Inspection Service, Apr. 17, 1987 447-3367). CRS-33 been involved in validating one commercial imunoassay. In addition, the Laboratory is negotiating research contracts to develop immunoassays for USDA and the Department of Defense (DOD).95 CONCLUSION It appears that the scope of coverage of existing multiresidue methods systematically limits the capability of FDA's Pesticide Monitoring Program to detect a significant number of pesticides with food residue potential. Some of the chemicals that are not detected by these methods include pesticides that FDA has ranked as having high priority for monitoring because of potential health risks. "New" rapid test methods that use biological reagents such as enzymes and antibodies appear to complement the capabilities of existing methods. If surface food residues can be established as a reliable indicator of the cholinesterase-inhibiting pesticide load of produce, and if the ChE bioassay can be validated, it would offer a relatively inexpensive method of identifying produce free of organophosphate and carbamate pesticides. The potential capabilities of immunoassay are greater and more diverse. Immunoassay appears to offer a relatively inexpensive, rapid, and sensitive single residue method or class-specific screening method. Although technical difficulties make the potential application to multiresidue analysis less certain, it is likely to be useful in testing samples for 12 or fewer compounds. In addition, it may be the method of choice to test for "new" large-molecule pesticides produced by genetic engineering. There appears to be widespread interest in bioassays and immunoassays. The usefulness of these methods cannot finally be determined unless they are 95 Williams, Llewellyn. Personal Communication. CRS-34 validated and compared to classical methods. In addition, it should be appreciated that significant developments have occurred in automating and otherwise speeding up analysis by classical methods. These developments are not addressed by this report. The capabilities of the Pesticide Monitoring Program may best be improved through a combination of these methods. Finally, if rapid test methods are eventually incorporated into the program, the impact is likely to be greater than merely adding a new technique. The methods could expand the role of screening in the monitoring program. Screening carries with it significant implications for program cost as well as program design. These issues cannot be thoroughly analyzed without a more detailed study of the rapid methods and their potential application in monitoring pesticide residues in food. LIBRARY ‘ ’ OF WASHINGTON UNIVERSITY 31'. LOUIS — M0.