Document OE29EN0ZXdrYBDxOgbzmmp0NM

PVC IN FOOD AND DRUG APPLICATIONS BRUCE H. BORSUK APRIL 12, 1985 VEU-144031 CONTENTS INTRODUCTION .......................................................................................................................................PAGE 1 PVC FOR FOOD CONTACT AND LIQUOR BOTTLES ........................................................................................................................ PAGE 2 GOOD MANUFACTURING PRACTICES ................................................................................................ PAGE 5 FDA APPROVALS AND THE MASTER FILE SYSTEM .........................................................................................................................................PAGE 7 PVC FOR PACKAGING DRUGS AND BIOLOGICS .................................................................................................................................... PAGE 9 PVC AS A COMPONENT OF MEDICAL DEVICES ............................................................................................................................... PAGE 10 PVC FOR COSMETICS PACKAGING ...................................................................................................PAGE 11 MASTER FILES: ESTABLISHMENT PROCEDURE, CONTENTS, AND OTHER MATTERS ...................................................................................................PAGE 12 1 * INTRODUCTION The regulatory status of PVC in the food and drug industry is, not surprisingly, a complicated issue. It is my general conclusion at this point that past and current regulatory inaction on the part of the FDA and other agencies has had as much impact as determing how and where PVC can be employed as any adverse regulatory action has had. A major regulatory proposal concerning the use of PVC in contact with food, for example, remains in limbo, neither finalized nor withdrawn. Vista, as a supplier of PVC, must abide not only by the regulations of the FDA, but also by the desires and interpretations of its prospective customers, who have their own standards and expectations regarding the current and future use of PVC in food contact applications. In spite of this discouraging description of the PVC regulatory scene as being somewhat less than concrete, there are certainly some definitive rules and procedures governing the use of PVC in many applications. Knowledge of these rules and procedures and adherence to appropriate regulation should enable us to pursue, and eventually sell to, customers who have eluded us in the past. As for the PVC applications which do not benefit from such regulatory direction, monitoring of forthcoming regulatory action as well as dialogue with prospective customers is advised prior to any major product or market development actions. 144033 2 PVC FOR FOOD CONTACT AND LIQUOR BOTTLES The Food and Drug Administration has jurisdiction over the use of PVC in food packaging, cosmetic packaging, drug packaging and as a component of medical devices, although the regulations differ for each of these areas of application. In 1958, the Federal Food, Drug and Cosmetics Act was amended by the addition of a Food Additives Amendment, which defines a food additive but provides that the term does not include "...any substance used in accordance with a sanction or approval granted prior to the enactment of this paragraph...," PVC is such a prior-sanctioned substance. Prior to 1958, PVC had been used in flexible food contact applications as well as for the production of some rigid food-containing bottles and was generally recognized as safe for these uses. More technically, PVC resin qualifies for prior sanctioning provided it has a maximum volatility of not over 3% when heated for one (1) hour at 105C, and an inherent viscosity of not less than 0.35. In 1968, industry urged the Bureau of Alcohol, Tobacco and Firearms (BATF) to test the suitability of using PVC for liquor bottles. Final approval depended upon satisfaction of FDA food safety criteria, but initial authorization was one step removed, and lay within the jurisdiction of BATF. Beginning in 1968, BATF permitted the marketing of PVC liquor bottles in specific sizes on an experimental basis. Over the next three years, industry worked closely with BATF in the hopes that the experimental tag would be removed from PVC liquor bottles. In 1971, the Department of Treasury (of which BATF is a part) decided that allowing full-scale marketing of PVC liquor bottles would be "a major federal action significantly affecting the quality of the human environment". It was the opinion of the Department of Treasury that such a move would require the preparation of an Environmental Impact Statement (EIS) in accordance with the National Environmental Policy Act of 1969. The EIS was Issued by BATF ^-14403* - 3- f in early 1973 and was very favorable to PVC manufacturers, concluding that FVC bottles are less energy intensive to produce than their glass counterparts. The GIS endorsed approval of PVC as a suitable container for use with distilled spirits. Just a BATF was finalizing its environmental assessment of PVC liquor bottles, Schenley Distillers, Inc. surprised both industry and government with its finding that trace amounts of vinyl chloride were migrating from PVC bottles to distilled spirits such as vodka and gin. This finding was confirmed by the FDA and others and on May 11, 1973 BATF, preempting a proposed FDA regulation prohibiting the packaging of alcoholic beverages in PVC bottles, terminated the experimental use of PVC packaging for distilled spirits. This FDA proposal, although precluding the use of PVC resin for packaging liquor bottles, would have permitted the continued use of PVC in other rigid and semi-rigid food applications. Based on data generated by OSHA concerning the toxic affects of exposure to vinyl chloride, however, the FDA in 1975 broadened its proposal and indicated an intent to ban all rigid and semi-rigid PVC products while allowing continued use of PVC for thin film and coatings or closure liners (Attachment A). This FDA proposal, never made final or withdrawn, forms the basis of the "little black cloud" which hangs over PVC. It is unlikely that this proposal will ever become law. The FDA, on October 15, 1979, even responded in a letter to GTR that "PVC continues to be prior sanctioned for food contact purposes providing that no vinyl chloride monomer migrates to the food." In this letter the FDA also indicates that, since the 1975 proposal was published, "Manufacturers have improved their manufacturing techniques so that little or no vinyl chloride remains in the polymer". On April 2, 1982 the FDA published an Advanced Notice of Proposed Rule Making (Attachment B) which discusses a "constituents policy" applicable to the PVC problem. It is expected that this policy will withdraw the 1975 YEU-l44035 I -4- proposal and clear the use of PVC products In contact with all foods subject to some limitation on either the residual monomer content or, more likely, the maximum concentration of vinyl chloride that may migrate. Constituents in this area refers to substances not intended to be either direct or indirect food additives, yet unavoidably present in the additive, such as residual VCM. It seems the FDA has admitted that it will accept the presence of a carcinogen as a constituent of a non-carcinogenic additive. According to Keller & Heckman attorney John Eldred, "Prudence suggests and most users require that PVC compounds Intended for use in food-contact application be made with the lowest feasible residual monomer levels". Both the PVC industry and BATF await this constituents policy. On September 28, 1984, Food and Drug Commissioner Frank E. Young stated in a letter that there is "no reason at the present time why a rule establishing safe conditions for use of PVC bottles and other food-contact uses of vinyl chloride polymers should not be Issued". In December of 1984, Jerome Heckman of Keller and Heckman wrote to Mr. Young endorsing that statement and urged the commission to issue "without delay" such a food additive regulation. As of this writing, this has not occurred. <JEV-144036 GOOD MANUFACTURING PRACTICES All direct or indirect-food additives, whether prior sanctioned or granted approval through specific regulation, must be produced using "good manufacturing practices". This term is dealt with, vaguely, in the Code of Federal Regulations, Part 174 (Attachment C). This regulation is surprisingly devoid of any references to specific manufacturing practices and contains such generalities as, "any substance used as a component of articles that contact food shall be of a purity suitable for Its Intended use". In September of 1983, Bill Percival of DuPont's Patents and Regulatory Affairs Department, in a response to an Inquiry from SKS, offered the following interpretations of the practical meaning of good manufacturing practices: a) When considering production facilities, keep in mind "reasonable precaution" and "prudence"; b) positive pressure enclosure for a compounding line would not be necessary; c) we should consider physical separation of extruders (for compound or dryblend production) using some sort of enclosure to prevent anything from dropping from ceiling beams or walls into the resin, separate storage of food-grade vs. non-FDA additives (or reactants), use of color codes on packages to designate food-grade vs. non-food grade additives. -6- > use of checklists, operating instructions, etc. to absolutely minimize any chance of operator error, - regular operator training on the importance of maintaining the food-grade product in pristine form. In the final analysis, however, a resin is considered to have been produced using good manufacturing practices if it meets the volatility (3.0%) and I.V. (0.35) specification mentioned previously. (See Attachment D) UEV-144038 7V FDA APPROVALS AND THE MASTER FILE SYSTEM In 1962 the Federal Food, Drug and Cosmetics Act was again amended by the passage of the New Drug Act Amendments. This new law gave the FDA the responsibility, for the first time, for passing on drug efficacy as well as safety. At about this time, plastics were gaining widespread popularity for use in drug packaging, especially bottles, and the FDA found it necessary to consider a drug's packaging when evaluating the drug's efficacy. In order for the FDA to accomplish Its evaluation as described above, it must be supplied with information concerning the drug as well as the intended packaging. Data pertaining to the drug itself, of course, comes from drug manufacturers. The drug manufacturers may also supply the information concerning the intended packaging material if such information was made available to it by the packaging suppliers. In order to keep producers of resin, compound, or bottles (in the case of PVC drug bottles for example) from having to share with their customers information generally considered trade secrets, the FDA set up a system which has come to be called Master Files. The FDA's Master Files system is completely voluntary and allows the FDA to collect and keep confidential all information necessary to pass judgement on the suitability of a particular drug's packaging. An SPI manual published in 1967, and still pertinent, states that plastics manufacturers "....usually provide FDA with lists of the components of their products, relevant details about the manufacturing process used, a description of the chemical composition and physical characteristics of their products, explanations of product designations used, and data concerning the percentage concentrations of components or formulations", for inclusion In a Master File. There are actually two types of Master Files; a Drug Master File (DMF) and a Medical Device Master File (MDMF). Each serves the same purpose and can contain similar, if not Identical, information. The difference being that VEV-144039 - 8- * the Drug Master File is referenced by the FDA in conjunction with a New Drug Application (NDA) or Investigational New Drug Exemption Application (IND), while the Medical Device Master File is referenced by the FDA in conjunction with a Medical Device Application, as required by the New Device Statute of 1976. VEV-14404Q -9- PVC FOR PACKAGING DRUGS AND BIOLOGICS As previously mentioned, containers used to package drugs are considered to be a part of the drug and, therefore, are cleared at the time the FDA issues its approval of a New Drug Application. Blood bags contain an anti-coagulant and are regulated as biologies. In both cases, it is the responsibility of the drug or biologies manufacturer to supply the FDA with information concerning its product's composition and packaging. This can best be accomplished through the Drug Master File system described above. The manufacture of resin, compound, and bottle (or flexible sheet in the case of the blood bag) may each set up DMF's at the FDA. These raw material manufacturers can then authorize the drug or biologies manufacturer to request the FDA to refer to the DMF in its review of the New Drug Application. These steps may also be followed when the drug or biologies manufacturer wishes to approve a new source of supply, although in this case the FDA's review will be cursory for a raw material such as PVC resin. It is important to the drug or biologies manufacturer that his raw material supplier have a DMF set up and be able to provide the manufacturer with i^s assigned DMF number. In the case of a New Drug Application, if the FDA needs further information relevant to the packaging, a request will be made of the appropriate suppliers. VEV-144041 10 - PVC AS A COMPONENT OF MEDICAL DEVICES As with drugs and biologies, the manufacturer of a medical device is required to register and list the devices it manufactures. If Vista Polymers produces a resin or compound and markets it (them) specifically for use as a component of a medical device, Vista would incur the registration and listing obligation. If, however, we produce resin or compound for general uses which may include its employment by a device manufacturer as a component of a medical device, the registration and listing requirement is not applicable. All medical devices are placed into one of these categories; Class I devices are subject to general controls applicable to all devices. Class II devices are subject to performance standards in addition to general controls. Class III devices are those which are intended for use in supporting or sustaining human life and are subject to premarket approval similar to that required of new drugs. These classifications are of no particular concern to us as suppliers of PVC, however potential customers sometimes state that our resin or compound must have "Class III certification", for example. Such certification for PVC does not officially exist, and any necessary approvals of our products would be handled by the FDA's Master Files procedure. The manufacturer of a device that wishes to employ a Vista PVC product as a component of a device may be required to provide FDA with information regarding the PVC component. Again, the device manufacturer, like the drug or biologies manufacturer, requests the FDA to reference Vista's Medical Device Master File identifying it by its Master File number. 144042 PVC FOR COSMETICS PACKAGING The FDA has no requirements for PVC compounds used to package cosmetics. Nevertheless, many manufacturers of shampoo as veil as mouthwash require that PVC compounds used to make bottles for their products have a very low residual monomer level, essentially the same as that requested for the manufacture of food-contact products. The FDA does not seem to have any special Interest in cosmetic containers and John Eldred of Keller & Heckman does not anticipate the FDA will direct any attention to this area when it releases the "constituents policy" discussed above. 12 - V MASTER FILES; ESTABLISHMENT PROCEDURE, CONTENTS, AND OTHER MATTERS Value to Vista The existence of a Drug Master File at the FDA Is a necessity for approval of our resln(s) at Klockner who, although supplying markets for which PVC Is already prior-sanctioned, require each resin supplier to communicate its DMF number as part of the resin approval process. In addition American Hoechst, who produces large volumes of pharmaceutical blister packaging using a 5265-type resin, will probably need a DMF number to approve our lowest molecular weight resin. Both a Drug Master File and a Medical Device Master File will allow marketing and sales to develop many accounts from which we have been excluded due to FDA issues as well as due to a customer's inability to segregate FDA-approved from non-FDA approved resins. (To clarify: having "FDA-approved" resin simply refers to having the appropriate Master File and adhering to good manufacturing practices. Actual approval cannot be granted until the IfDA evaluates a particular resin in combination with an inten4ed use as evidenced by a customer's New Drug Application, Medical Device Registration, or notification of new source of supply.) Although the medical applications market for PVC is somewhat smaller and more fragmented than our traditional markets, recent market research confirming its high growth potential, as well as the opportunity for greater-than-pipe netbacks make it a promising target for our non-pipe diversification strategy. Furthermore, the steps necessary to position our products for FDA related applications (l.e. establishment of FDA Master Files) will likely require no capital expenditure or product modification beyond that which we are already considering for the rigid calendering market. An evaluation of the FDA status of our reactor additives will be necessary, however. VEV-144044 13 - * Just a few potential accounts requiring "FDA-approved" resin are Johnson & Johnson (purchase 1.5-2.0MM lbs./yr. of resin for production of band-aids and medical tapes); Baxter-Travenol; Vinylex; Sherwood Medical; and various companies who produce compounds specifically for FDA applications. Also, consider the possible value of a DMF when recalling the anticipated release of FDA standards for use of PVC in the manufacture of liquor bottles. I recommend we begin immediately the steps necessary to establish both a DMF and an MDMF. Number of Files We may include information on any number of our five resins, or all five, in a DMF, the same for an MDMF, so a maximum of two FDA files will be necessary for resin. As for a Vista bottle compound, dryblend, or pellet, 1 suggest a separate DMF or MDMF in which we reference our own resin Master Files. Contents of Files The FDA advises we include at least: 1. Formulation 2. Toxicity data 3. Other applicable test data 4. Index of material submitted for the file Jerome Heckman has advised us to include some truly confidential information as this will be necessary for the FDA to establish a file for us in the face of storage and filing space limitations. He suggests: complete formulation including the identity of the catalyst (initiator), surfactants used (if any), and molecular weight control agents (if any). UEV-144045 f 14 - V If the FDA, when evaluating our file in conjunction with a customer's application or notification, requires further data such as results of extraction studies, we may add it upon their request. In general, the contents of our file can always be expanded or updated as our business dictates. Confidentiality of File Contents The FDA maintains that the contents of the files are confidential, however questions have arisen regarding the applicability of the Freedom of Information Act (FOIA). The FOIA provides specific exemptions which allow governmental agencies (such as FDA) to keep certain documents confidential. Exemption 4 of the FOIA concerns "documents containing trade secrets of commercial or financial information obtained from a person outside government and privileged or confidential". (Interestingly, a case known as Continental Oil Co. vs. Fuel Products Commission in 1975 is cited as a relevant test case of this exemption.) "Trade secret" has been defined in the courts as "... a secret, commercially viable plan, formula, process, or device that is used for making, preparing, compounding or processing trade commodities and that can be said to be product of either Innovation or substantial effort." "Confidential information" is that, as to which, "disclosure would cause substantial harm to the competitive position of the party from whom the information was obtained". An FDA Master File would qualify for this exemption. Cost The FDA reports there is no fee charged for the establishment of a Master File. If we use the services of Keller & Heckman, charges would be "about $500.00" according to Peter de la Cruz of that firm. Keller & Heckman serves to minimize delays by hand-delivering our letter and information to the FDA, and following up "as frequently as may be necessary" until the Master File number is assigned. What FDA Does Initially FDA simply assigns a number to the DMF or MDMF and communicates this number to us. It is purely an administrative task which will take anywhere from "a couple of weeks" according to a recent conversation with FDA, to "A months" according to Mr. de la Cruz of Keller & Heckman, referring to a current backlog of work at FDA. As alluded to previously, the FDA does not "approve" the material submitted as they do not know in what type of applications the resin will be employed. As for the probability of our resin gaining approval for the applications we are already considering, I suggest we first consult the suppliers of our antioxidant, CTA, etc. as they should be able to review with us the FDA status of their products. Concurrent with this we can ask FDA for a preliminary review of the suitability of our resin for several named applications in our request for Master File establishment. Miscellaneous For both DMF and MDMF, one original and one duplicate original of all material must be submitted. The DMF is sent to the FDA's Drugs & Biologies Product Coordination Center; the MDMF to the Center for Devices and Radiological Health. \440A7 ATTACHMENTS A Proposed FDA Regulation prohibiting use of PVC for production of rigid and semi-rigid food-contact articles. B Advance Notice of Proposed Rule Making discussing a policy for regulating constituents of direct and indirect food additives. C Part 174 - Indirect Food Additives: General including definition of Good Manufacturing Practices. D SPI Manual "Good Manufacturing Practices" Criteria for Plastic resins "Prior Sanctioned" under the Food Additives Amendment of 1958. VEV-144048 t ATTACHMENT A VEM-144049 PROPOSED RULE5 (2) Long varieties. U S. No. 2. or bet ter err.de, l7a inches minimum diameter. (3) All varieties. Site B, if US. No. 1, or better grade. (b) Maturity a!, inning) require ments. <1) Russet Burbank and Red McClure varieties. For U.S. No. 2 grade not more than "moderately skinned" and for other grades not more than "slightly skinned," (2) All other varieties. Not more than "moderately skinned." <c> Inspection. tl> No handler shall handle any potatoes for which inspection is required unless an appropriate inspec tion certificate has been issued with re spect thereto and the certificate is valid at the time of shipment. For purposes of operation under this part it is hereby de termined pursuant to paragraph (d) of 5 948.40, that each inspection certificate shall be valid for a period not to exceed 5 days following the date of Inspection as shown on the Inspection certificate. (2) No handler may transport or cause the transportation by motor vehicle of any shipment of potatoes for which an inspection certificate is required unless each shipment is accompanied by a copy of the inspection certificate applicable thereto and the copy is made available for examination at any time upon request. <d) Special purpose shipments. (1) The grade, size, maturity and inspection requirements of paragraphs (a>, (b) and (c) of this section and the assessment requirements of this part shall not be applicable to shipments of potatoes for: (1) Livestock feed: (ii) Relief or charity; or flii) Canning, freezing, and "other processing" as hereinafter defined. (2) The grade, size, maturity and In spection requirements of paragraphs (a), (b) and <c) of this section shall not be applicable to shipments of seed pur suant to | 948.6 but such shipments shall be subject to assessments. (e) Safeguards. Each handler of po tatoes which do not meet the grade, size, and maturity requirements of para graphs <a) and <bi of this section and which are handled pursuant to para graph (d) for any of the special purposes set forth therein shall. (1) Prior to handling, apply for and obtain a Certificate of Privilege from the committee. '2) Furnish the committee such re ports and documents as requested. In cluding certification by the buyer or re ceiver as to the use of such potatoes, and (31 Bill each shipment directly to the applicable processor or receiver. (f) Minimum quantity. For purposes of regulation under this part, each per son may handle up to but not to exceed 1,000 pounds of potatoes without regard to the rcquiiciuents of paragraphs (a), (b) and <c> ot this section, but this ex ception shall not apple to anv shipment which exceeds 1,000 pounds of potatoes. <g) Definitions. The terms "US. No, 1," "U.S. No. 2," "slightly skinned," and "moderately skinned" shall have the same meaning as whin ti ed in tile US. Standards for Potatoes <55 51.1540-51 1566 of this title, effective September 1, 1071, as amended), Including the toler ances set forth therein. The term "other processing" has the same meaning as the term appearing in the act and includes, but is not restricted to, potatoes for de hydration, chips, shoestrings, starch, and flour. It Includes only that preparation of potatoes for market which Involves the application of heat or cold to such an extent that the natural form or stability of the commodity undergoes a substan tial change. The act of peeling, cooling, slicing, dicing, or applying material to prevent oxidation does not constitute "other processing." Other terms used in this section shall have the same mean ing as when used in Marketing Agree ment No. 97, as amended, and this part. th> Applicability to imports. Pursuant to section 8e ot the act and 5 980.1, Im port regulations (7 CFR 980.1), Irish potatoes of the red skinned round type, except certified seed potatoes. Imported Into the United States during the period September 29, 1975, through June 30, 1976, shall meet the grade, size, and quality requirements specified in para graph (a) of this section, and during the effective date ot this handling regula tion through October 31, 1975, shall be not more than "moderately skinned," Dated: August28,1975. Charles R. Brader, Deputy Director, Fruit and Veg etable Division, Agricultural Marketing Service. [FR Doc.75-23304 Filed 9-2-76;8.-48 m| DEPARTMENT OF HEALTH. EDUCATION. AND WELFARE Food and Drug Administration [ 21 CFR Part 121 ] [Docket No. 75N-0235) PRIOR-SANCTIONED POLYVINYL CHLORIDE RESIN Termination of Notice of Proposed Rulemaking posed regulations concerning the u-e of vinyl chloride polymers m contact with all types of food. Accordingly, the Commissioner hereby withdraws the proposal published on May 17, 1973, and terminates the rule making proceeding beam by that proposal. This action is taken pursuant to tile provisions of the Federal Food. Drue, and Cosmetic Act (secs 4ft9, 701(a), 52 Stat 1055, 72 Suit. 1785-1788 (21 U.S.C. 343. 371<ai) and under the authority dele gated to the ComtiUs-inner ol Food and Drugs 121 CFR 2.1201, Effective date. This order shal! Lc ol feclive September 3. 1975. Dated: August 27, 1975. A. M. SCHMID7 Commissioner o; Food and Drugs |FR Doc.73' 23240 Filed 8 25-75.8 (Sam| [ 21 CFR Part 121 ] {Docket No. 75N-0190) VINYL CHLORIDE POLYMERS IN CONTACT WITH FOOD Notice of Proposed Rulemaking JAe. ?d and Prug Administration (FDA)Uroosingregjilationsjp^^ ^mctjiiejjs^-oi_vinjTcnlonaejoivmers In contact with iooq. me proposal permits the continued use of \nvl mil lofld-contact articles where the potHjtlalformigratio^Mvimqctuortd^ Jg,ajgyjnsneaj*rtlie_e)^^ pot reasonably be expected to become a ^gjjjgojjgjj^pMood. The proposal includes an mterimrooTadditlve regulation for the use of water pipe made from vinyl chloride polymers. The Interim regula tion would be in effect pending develop ment of additional data to determine if vinyl chloride may reasonably be ex pected to be in potable water that is drawn from the tape. ^Jjyjrggggg^rggjj- ____ Interested persons have tint November 3. 1975 to submit comments Use or Vinyl Chloride Vinyl chloride is a chemical with the following structure: nn \, / It has a boiling point of -14* C '6.8" F) and consequently it is ordinarily a gas. This property led to its use as a propel lant for aerosol products such as cosmet ics, drugs, and pesticides. Vinyl chloride Ls polymerized to form polyvinyl chloride in which the basic monomeric unit i> i lpeated: - n ii - --r c - L ui n1 J,, In this formula, n represents the num ber of monomeric units that may be present, a sum which normally exceeds 800 units. Vinyl chloride homopolymers and co polymers are used In the production of articles or components of articles In- FEDEPAl REGISTER, VOL 40, NO. 171--WEDNESDAY. SEPTEMBER 3. 19,-5 T in.vio PROPOSED RULES 'crc'.v''. : ? contact load, includ':!^ foocl- -r~, in- in, ar. ticlies for contact with the remainin_g__u__py! t:.i- _ r.pv hp pr.ckr.-o'" material.-,. coatings. pa:t? for alcoholic loods and was based on: (1) VCCl in tri1? n:> prrni: -n ^>.1.111! load processing equipment, flexible tub in? rmcling innTTSTBnSrVnBnTTToride chloytle "comno.imi lor such Ht-CTTF ing. and water pipes. in polyvinyl cliloride bottles was being Plastic ms Uriah cs mini. ana Polyvinv^rhlorid^bassevrivljTraji; extracted by bottled distilled spirits and ling. In n>rrc< ;;nithf ''com; oiuuu" wines, and <2) the fact that no available 10 polyvinyl clil*: it :n ;x iniM-.d and sun i| ^as ^inntv. to aninv'.l feeding studies established a safe heated with rr- v.ufh a- :'a to ra .-arts of level of coiv-umption when vinyl chloride plasticizer per l.-g-drc-l was extracted from containers Into food, The addition <! Hie phi "ik-ver 1 opens" At that time there were no data that in tho spaces between the pclvmer dicated that polyvinyl chloride articles "threads," and tho heat lends lo drive in contact with nonalcoholic foods out the remunmg vinvl chloride mole cliloride are u=ed cm it year in the pack vculd result In migration of residual cules. Fui ther oppoi hm:1 s' for removal of aging of food, moling it 'after j^oly- vinyl chloride Into the food and, accord vinyl chloride is prov:.;>d dJiim: th.e ethylene i the strand most commonly used plastic for packaging food. The pro duction of water pipe is one of the single largest uses of polyvinyl chloride, ac counting aimuativ for ov'-r <!f'(> million pounds of resin. In addition, UliJOull- 2^vinri^hloridc2invalsobcejj (i*bl)(-!)TuT_a__^Bcliaglng m_a_t_e_ria^^or .m.w.M.K*.nt>nnaihiEMinraitB on. ingly. there was no reason to consider rc'-tvtct ton of such uses. Tho exigence of residual vinyl chlo ride in articles made from vinyl chloride polymers is related to the manufactur ing process and the physical structure of the polymers. The gas vinyl chloride be comes the packaging material consisting of polyvinyl chloride or one of the co fabrics lion of the "comt ound" into ar- tides because heat is urr.ally need in the dpirfofciceu:Hlt.ifnBoir:v.tthhceonvtii,n--y,tl.cch!1loric`5Temupc-rhKmmmol?srlAee ,ttoy escape from rigid and scecmmi:irilggiidd cco?llvv~vinvl chloride articles beemre tiny coatalp. litUa 3T jecr and nr.n.iJ'.tjrranTg-.nTaiMoimr.igmi polymers of vinyl chloride through a se metjes. Vinn chloricie.polymers arejstso used as components or certain medical gerices. Early investigatory work reviewed by Dr. A. J. Lehman, then Chief of the Division of Pharmacology, Food and Drug Administration, indicated that polyvinyl chloride was insoluble In vari ous solvent systems used to simulate food. Consequently, there was little con cern about the safety of food-contact articles made from prlwinvl chloride. In 1950. Dr. Lehman reported to the As sociation of Food nnd Drug Officials: XVo coa-'.der as the most Important slnele physical characteristic of a film Its solubility or the leaching out of any of Its constituents in the common media with which the plastic may come In contact Tf nothing con be extracted when tested with representative food-type solvents (lard-oil. vinegar, sodium bicarbonate, meat juice, water, etc.) under conditions somewhat more rlporoua than might be experienced under practical usage, we usually have no objections to use of tho film In situations where direct contact with food may result Toxicological problems re late more to the plasticizers employed to y'.vc tho film pertain desirable character ises than to the film Itself Plasticizers are legion, but to develop one which Is nontoxic and yet efficient is not easy of accomplish ment. Reports of this work were published by Dr. Lehman In "Chemicals in Food: A report to the Association of Food and Drug Officials on Current Developments," Quarterly Bulletin of the Association of Food and Drug Officials of the United States, 15(3) :82, 1951. and "Food Pack aging." Ibid., 20(4) :139, 1956. Migration or Vinyl Chloride to Food ries of distinct steps. The first step Is the polymerization of the vinyl chloride to form a polymer, e.g,, polyvinyl chloride resin. This resin is then blended with a number of other substances that may in clude plasticizers, stabilizers, lubricants, and processing aids to form a compound ed material ordinarily referred to as a "compound," e.g.. a polyvinyl chloride flexible film "compound" or a polyvinyl chloride bottle "compound." This "com pound" Is then used by the fabricator to produce the finished article that is used in contact with food. The individual molecules of polyvinyl chloride may be visualized as short strands of thread. The individual mole cules are attracted to each other by phys ical forces that tend to hold them to gether so strongly that the polyvinyl chloride, by itself. Is rigid. The polymeric material contains an exceedingly large number of polyvinyl chloride molecules, which are Intermingled and provide a number of open spaces (interstices) among the Individual molecules. The origin of the possibility that vinyl chloride may migrate to food Is its In complete polymerization into polyvinyl cliloride. Estimates indicate that some what less than 90 percent of the vinyl chloride Is converted to polyvinyl chlo ride. Most of the remaining 10 percent vinyl chloride is either vented to the at mosphere, or recovered by techniques, such as vacuum stripping, and reused. However, some vinyl chloride remains In the polyvinyl chloride resin following Its polymerization and Isolation: time as much as 2000 Darts Der million ffipmTTeslauar^nyrcElofHeremami ual vinyl chloride. JjowgvgrJjTJAhas not bb^eeeqn able to detect jvininvvllchcnlolaflfdlgepmH of the plastlclzec. flexible polyvinyl emonae products it has analyzed Even In the case of rigid unplasticized polyvinyl chloride, there is a loss of vinyl chloride during fabrication into ar ticles. Once fabricated, the polyvinyl chloride articles continue to lose vinyl chloride by diffusion. Data have been developed showing a gradual loss of re sidual vinyl chloride from polyvinyl chloride articles during their storage prior to use. This diffusion phenomenon continues to occur when the polyvinyl chloride is used in contact with food. If vinyl chloride is present, a certain amount may be expected to migrate to the food. In the case of semirigid and rigid articles that contain high levels of residual vinyl chloride, this amount has been shown to be substantial. However, in the case of coatings, films, and other plasticized food-packaging materials in which the amount of residual vinvl chlo ride is extremely small, there appears to be little likelihood that vinvl chloride would reasonably be expected to be pres ent In the food. The migration of vinyl chloride may be viewed as a simple diffusion phenom enon: The vinyl chloride is leaving the location of highest concentration, the plastic article, and moving to a lo cation of lower concentration, whether It be the surrounding air or the food contained in the arttcle. This hypothesis appears to be supported by the work of scientists at Ethyl Corp. ("VCM extrac tion from FVC bottles," Modern Packag ing. pp. 45-48, April 1975). Their data Kfh imir,T?d .ltI5r4l7dFHTiHi*li'lifK, were subsequently con-i-l-t-i-i-l&-- Vi^sW *!' MliMBi PV>Wh<$ ft ffit FltPtW FfM; as il to z ppm remain, it is theorized Hi Vlfiyl iHlAflfllTiecomes physi cally trapped among the Interstices of the polymer "threads." This model permits an explanation of the varying degrees to which vinyl chlo ride is removed from the different forms of the polymer. A large amount of the vinyl chloride that falls to polymerize never becomes trapped in the resin; it either finds its way out of the resin maze, Indicate that the vinyl chloride levels In the plastic article and Its food content eventually reach a point of equilibrium. However, there continues to be a Ids': of residual vinyl chloride to the surround ing atmosphere so that the level in the plastic article becomes lower relative to the level In the food inside it. When this occurs, there is a migration of vinyl chloride from the food into the plastic article: This Is represented by a decrease ieb of Mav 17.1973 (38 FR 12931). would due to its volatility, or can be removed In the concentration of vinyl chloride have precluded use of polyvinyl chloride by vacuum stripping. Moat, if not all, of that can be detected in the food. The ob- FEDfDAl REGISTER, VOl. 40, NO. 1/1--WEDNESDAY, SEPTEMBER 3, 19/S veg-144051 PR POSED RULES io:>3i vious end point indicated by this hypoth esis is that there would be no vinyl chloride in the food or in the plastic article at that distant point in time when it has all migrated to the surrounding atmosphere. A modification and extension of this hypothesis has been proposed by Pro fessor Seymour Gilbert, Fh.D., Depart percent ethanol) and In polyvinyl clilo- rlde as low as 0.35 ppm. On December 20, 1973. representatives of FDA and the Society of the Plastics Industry, Inc. (SPI) met to discuss chemical Information concerning vinyl chloride that SPI had submitted with Us comments of October 15. 1973, on the original proposal of May 17. 1973. In re migration from film, cap liners, coat ings. gaskets or flexible tubing, suits of analyses ot extractives from such articles have shown no detectable vinyl chloride. Analyses of these plastic articlcs themselves have shown no detecta ble vinyl chloride using analytical meth ods reported to be capable of detecting a level as low as 1 ppm residual vinyl ment of Pood Science. Rutgers Univer sponse to questions raised at tills and chloride. No residual vlnvl chloride was sity. New Brunswick, N.J. Dr, Gilbert's subsequent meetings, members of the .found in FDA analysis of polyvinyl chloT SPI have obtained Information about -ride blood bags anndd, flexible tubingg__uussii?nig 1 n' rTcH t>3yrtnyf chloride tnaV findJo polyvinyl chloride, Including analyses of a method capable of detecting 0.35 ppm various types of foods for the presence residual vinyl chloride. molecules '1 ' -r--- -- of vinyl chloride, refinement of the The lack of findings of extractable :entrations methodology for detection, and review of vinyl chloride from film Is not surprising, v a vprv nidjt^ t the toxicological aspects. Memoranda of these meetings and the Information sup for theoretical calculations Indicate that If film contained residual vinyl chloride, ! polyvinyl crnonoe in ac- plied are on public display at the office of 100 percent migration of the residual the usual diHuslon theory. the Hearing Clerk, Food and Drug Ad vinyl chloride from a 1 mil (0.001 Inch) ministration, Rm. 4-65. 5600 Fishers film would result in 2 ppb vinyl chloride Lane, Rockville, MD 20852. In food. These calculations assume that Data supplied by SPI indicated that 10 grams of food contact each square vinyl chloride could migrate to nonalco Inch of film, the film weighs 20 milli holic foods from polyvinyl chloride bot grams per square Inch per mil thickness, _ 41s theory is supported Dy the tles. Analyses of two samples of vegetable and the film contains 1 ppm residual results "of equilbrlum studies In which oils packed In polyvinyl chloride bottles vinyl chloride. However, these assump powdered polyvinyl chloride resins con revealed the presence of vinyl chloride tions are exaggerated, e.g., the vinyl chlo taining no vinyl chloride were added to at levels of 1.6 and 6.5 ppm. Reported re ride will migrate Into the air as well as food-simulating solvents containing known concentrations of vinyl chloride. Vinyl chloride was found to be taken up by the resin and its concentrations in the food-simulating solvents were reduced to a much greater; extent than would be ex sults of analyses of additional samples of foods, drugs, and cosmetics disclosed varying levels of vinyl chloride, e.g., vine gar (5 ppb), mineral oil (74 ppb and ap proximately 2 ppm from two other sam ples a vitamin supplement (approxi the food, and much lower levels of migra tion into food would be expected to occur under actual conditions of use. to the point where they would be extremely small. Tlie data substantiate this con clusion because the levels of extractable plained by simple diffusion or partition ing. Schenley Distillers reported levels of vlnjT c!rlo?ide as high as*!Q ppmJnViflR mately 1 ppm), and a mouthwash (174 ppb'. No vinyl chloride was reported from a series of water samples using an analytical method reportedly sensitive vinyl chloride have never been shown to approach 100 percent In those cases where actual values have been presented for article thickness, residual vinyl chlo ^jj^zsjjjmr^niL Confirmatory work, on to 50 ppb. These latter samples had been ride level, and levels of extraction of vinyl samples of the same material by FDA collected from operating potable water chloride. showed levels of 11 ppin vinyl chloride systems of various polyvinyl chloride for In vodka and 12 ppm in gin. Since tha{t mulations at six different building sites. time FDA lias received many additional In early 1974, the British Ministry The greatest likelihood for reports or findings or vlnvl chloride in of Agriculture, Fisheries, and Food re ported finding vinyl chloride In con centrations ranging from 10 to 80 ppb in reports _ __________ . ___ matlon to evaluate accuracy, such as: an adequate description of the metnoiT ology. including chromotograms: data frrom reco..v..e..r.y "s'tud"ies verif'ying the fruit squashes and from 10 to 40 ppb in cooking oils. The Canadian Health Protection Branch Of the Ministry of Health and Welfare reported finding vinyl chloride ranging from 0.9 to 8 4 aged In such containers, including vege table oils, vinegar, honey, and liquid vitamin supplements. Large Quantities a:Of Processed meats are packaged in rigid a-nnd seommiirriigfid c-o..n...t.a..i.n. ers composed "of Iaimed sensitivities; data showing con- ppm In seven samples of apple cider rmat_l_o_n__ _D_y_ m..a.ss spi -pcctrosr:copayn.d - vinegar. It also reported vinyl chloride In dentiflcationoOhe plastic material. It samples of various wines (less than 0.025 should be noted that analysis for vinyl to 0.98 ppm), gin (0.22 to 0.7 ppm), and chloride requires careful analytical tech malt vinegar (1.5 ppm). fvlUinlvclr ccholonrdidime epnotlsymaerers*.jJreeoilvu.ehnotlnveyp, aan~d eaed In , individual serving containers niques to assure credible findings. The Data from experimental work using analysis becomes progressively more difficult as the concentration of vinyl chloride decreases. The problems Involved with the anal ysis for vinyl chloride are emphasised by the difficulties that firms have en countered in obtaining consistent results during "round-robin'' studies in which a number of laboratories h ue analyzed the same material. FDA has developed a method for the detcur,motion of vinyl Chloride in polyvinyl chloride and In food-simulating rohent.i. 'Copies are available from the Dr.i-mn of Fuud and Color Additives. Food and Drug Admin istration, 200 C St., S\Y.. Washington, D.C. 20204.) Tins method is considered to be capable of measurin' levels of vinyl chloride In food-simulating solvents as food-simulating solvents, i.e., distilled water. 3 percent acetic acid, ethanol, and n-heptane, support these earlier data showing migration of vinyl cliloiide to types of foods other than alcoholic bev erages: The data show that the use of alcohol and n-heptane as solvents rep resenting alcoholic and fatly foods, re spectively. results in the highest levels of vinyl chloride extractives as compared to the amounts extracted by disLilled wa ter (representing aqueou1. foods) and 3 percent acetic acid (representing acidic foods). All of these data are on public display in the office of the Hearing Clerk. Food and Drug Administration. Tile available data indicate that cer tam applications ol vinyl chloride do mi nt~a rea"lisitic possibility of' vinyl chloride migration. The fcornmlssibner Is to 12 mils were reported In a submission from the American Meat Institute to contain 0 to 180 ppm lesiduol vinvl chlo ride and to yield 4 to 20 rr-b vinyl chloride when extracted by n-lieptane. Rigid articles with a thickness of 10 to 21 mils were reported by the American Meat Institute to contain 6 to 127 ppm residual vinyl chloride and to yield 2 to 237 ppb vinyl chloride when >d ruci-.d by n-heptane. Water pipe is a u e of polyvinyl chlo ride that presents little likelihood that vinyl chloride will become a component of potable water. The pipe's rigid, rela tively thick wall would be expected to have a potential for high levels cf resid ual vinyl chloride: data show lhat the level of residual vinyl chloride attainable low as 20 parts per billion <ppb> (In 50 unaware of any findings of vinyl chloride In water pipe may vary from less than FEDtRAl REGISTER, VOL. 40, NO. 171--WEDNESDAY, SEPTEMBER 3, 1975 144052 A 10.VJ2 PROPOSED RULES 10 ppm to more thnn 100 ppm. However, the potential lor extraction of viavl chlo ride from potable water pipe is greatly [educed because of the low solubility of vinyl chloride in water, the short time of contact, the large volume of water in contact with the pipe, and the compara tively low temperatures of exposure. The primary use of polyvinyl chloride potable water pipe is from water mains to build ings where a large volume of water flow occurs, and the temperature of exposure is lowered because the pipe is buried. Moreover, the small amount of vinyl chloride that might migrate into water from water pipe would be expected to dissipate during the aeration and agita tion that occur at the tap. Toxicity or Vinyl Chloride PDA Is unaware of any suitable tox iclty data from animal feeding studies that demonstrate a safe level of inges tion of vinyl chloride. Results of a 90-day feeding study with rats were submitted to FDA In October 1974. However, that study was not conclusive and, further more. could not resolve the prime Issue of safety, namely carcinogenicity, since it was a short-term feeding study. Life. time studies are necessary to evaluate properly the potential for long-range effects, such as carcinogenicity. Considerable data exist concerning the toxic effects of vinyl chloride from at mospheric exposures, especially by in halation and occupational contact. As the Commissioner pointed out in his April 22, 1974, proposal (39 FR 14215) to pro hibit the use of vinyl chloride as an in gredient in drug and cosmetic aerosol products: There is ample evidence that vinyl chloride inhalation can result In acuta toxicity mani fested by an arTay of symptoms, including unconsciousness as a result of high concen tration by Inhalation. Cardiac effects, bona changes, and degenerative changes In the brain, liver, and kidneys have also been re ported In animals. The results from studies by Torkelson, et al., on the chronic effects of vinyl chloride on laboratory animals <T. R. Torkelson, F. Oyen, and V. K. Rowe, "The Toxicity of Vinyl Chloride as Determined by Repeated Exposure of Laboratory Animals," American Industrial Hygiene Association Journal, 22(5), 354-361, 1961) Indicated slight effects in rats ex posed to atmospheres containing 100 and 200 ppm vinyl chloride. An exposure of 50 ppm was considered to be a "no-effect" level. Published scientific reports im plicate vinyl chloride as a causative agent for "acroosteolysls" of the hands and feet as well as systemic effects among industrial workers engaged in the manu facture of vtnyl chloride. Dr. P. L. Viola, In studies, exposed rats to an atmosphere containing 3 percent (30,000 ppm) vinyl chloride vapors for 4 hours per day for 5 days per week for 1 year (P. L. Viola, A. Blgottl, and A. Caputo, "Oncogenic Response of Rat Skin, Lungs, and Bones to Vinyl Chloride." Cancer Research, 31: 616-522, May 1971). He reported that rats subjected to such exposure developed tumors of the skin, lungs, and bones. Copies of these reports are on file with the Henring Clerk, Food and Drug Ad ministration. Reporting at the February 15. 1974. fact-finding hearing, which was called by a notice that the Occupational Safety and Health Administration published in the Federal Register of January 30, 1974 (39 FR 3874), Dr. Cesare Maltoni discussed preliminary results from his Investigations directed at clarifying the type and degree of carcinogenic effects of vinyl chloride, as previously reported by Dr. Viola. Dr. Maltani's Investigations In volved various types and levels of ex posure to vinyl chloride, Including: (1) An attempt to reproduce the conditions of Dr. Viola's experiment using a level of 30,000 ppm: (2) experiments using at mospheric exposure to vinyl chloride vapors at levels ranging from 50 to 10.000 ppm; (3) an experiment Investigating the effects upon Ingestion (intubation) of vinyl chloride: and (4) experiments Investigating endoperitoneal and sub cutaneous routes of administration. (C. Maltoni ti O. Lefemlne: "Carcinogenicity Bio-assays of Vinyl Chloride," Environ mental Research, 7:387-405, 1974 and "Le potenzialitA del saggl spertmentali nella predizione del rlschl oncogenl amblentoll. Un esempio 11 cloruro dl vinile," Accademia Nationals Dei Lincei, 56:1-11, 1974). In addition to rats. Dr. Maltoni reported that experiments were also being conducted using mice and hamsters. At the February meeting, Dr. Maltoni discussed his preliminary findings of the development of angiosarcoma of the liver, along with other types of tumors, at levels of atmospheric exposure as low as 250 ppm. At the New York Academy of Sciences meeting. May 10-11, 1974 ("Carcinogenicity Bloassays of Vinyl Chloride: Current Results." Annals of the New York Academy of Sciences, 246:195-218, January 31, 1975), he sub sequently reported the development of angiosarcoma of the liver and other types of tumors at levels of atmospheric exposure as low as 50 ppm. Further, he announced that additional experiments were being started; the experiments are using larger numbers of animals and lower dose levels of Inhalation exposure. Inhalation exposure studies using similar low levels of vinyl chloride are also In progress at Industrial Blotest Research Laboratories (IBRD under the sponsor ship of the Manufacturing Chemists As sociation. In discussing these two studies in regulations, published In the Federal Recister of October 4. 1974 (39 FR 35890), establishing standards for In dustrial exposure to vinyl chloride, the Occupational Safety and Health Admin istration stated: These investigators have Induced anglesarcoma of the liver In rats and mice at ex posure concentrations of 50 ppm and in hamsters at higher concentrations of ex posure. Additional tumors involving other orgRos. Including the kidneys, lungs, and skin of exposed animals, were also observed in frequencies much in excess of control animals. As noted above, the Food and Drug Administration Issued a proposal on April 22.1974, concerning the use of vinyl chloride as a propellant in aerosol drugs and cosmetics. At the same time, manu facturers were requested to recall any outstanding stocks of such products from tile market. A final regulation was pub lished in the Federal Recister of Au gust 26, 1974 (39 FR 30830). prohibiting the use of vinyl chloride as a propellant in cosmetic aerosols and requiring an ap proved new drug application for the mar keting of aerosol drugs containing vinyl chloride as a propellant. In separate actions, the Environmental Protection Agency, banned the use of vinyl chloride as a propellant in certain pesticide aerosols by notice published in the Federal Register of April 26. 1974 (39 FR 14753), and the Consumer Prod uct Safety Commission banned the use of other self-pressurized household products containing vinyl chloride, by a notice published in the Federal Register of August 21. 1974 (39 FR 30112). Dr. Cesare Maltoni has Issued a pre liminary report concerning the progress of his studies Investigating the effects of vinyl chloride when Ingested (Cesare Maltoni, Adriano Cilibertl, Luciano Gi anni. Pasquale Chleco, "Insorgenza Di Anglosarcotnl In Rattl, in Sequito A Sommlnistrazione Per Via Orale Di Cloruro Dl Vinile," GZi Ospedali della Vita, Anno II, Numero 1, Gennalo-Febbralo 1975). Dr. Maltoni's study involves the admin istration to rats by Intubation of vinyl chloride in an olive oil solution at dos age levels of 50 milligrams per kilogram of body weight, 16.5 milligrams per kilo gram of body weight and 3.3 milligrams per kilogram of body weight. The study was Initiated with 40 male and 40 female rats at each dosage level, plus a control group of the same number. After 52 weeks, the examination of those rats that had died revealed one rat tn the highest dose group to have angiosarcoma of the thymus, and a rat in the 16.5 milligrams dose level was found to have angiosar coma of the liver. No tumors were re ported tn the 3.3 milligrams dosage group or in the controls. The experiment is con tinuing with an anticipated completion date in early 1976. In addition, Dr. Mal toni has Initiated an experiment using lower dosage levels. After evaluating all the data, the Com missioner concludes that It is likely that when tha Maltoni study has been com pleted, It will show that vinyl chloride is carcinogenic when ingested. He notes that these results are consistent with the finding that Inhalation of vinyl chloride has been shown to produce cancer. The Commissioner acknowledges that the finding of angiosarcoma in one rat in each of the two highest dosage levels may be regarded by some persons as Inconclu sive evidence that vinyl chloride Is car cinogenic when Ingested. However. Dr. Maltoni reports that, to his knowledge, no spontaneous angiosarcomas of rats have been reported In the literature. Ad ditionally, Dr. Maltoni reports that an giosarcoma of the thymus and of the liver have never occurred spontaneously in their colony of Sprague-Dawley rats. The Commissioner concludes that the preliminary data from the incomplete In gestion studies, when combined with the FEDERAL REGISTER, VOL. 40, N . 171--WEDNESDAY, SEPTEMBER 3. 197S VEV-144053 A PROPOSED RULES J05.T! her data already available concerning ,e hazards of vinyl chloride, are suffi,ent to warrant tlie actions proposed in this proposal. pRoroscD Action lihood of the migration of vinyl chloride to food. Although testing for the car* cinogenlclty of vinyl chloride upon In gestion is not complete, the Commissioner concludes, as discussed above, that suffi cient data have been accumulated to es film because of the method of manu facture and the thickness of most film used for wrapping food (approximately f mil). As discussed above., Dpliassticizing the JUm. results in an article essentially free' of vtnvl chloride, and there is no reason Under section 201(a) of the Federal tablish the likelihood that it will be able expectation th.it any remaining Food, Drug, and Cosmetic Act (21 U.S.C. shown to be carcinogenic and therefore vinyl chloride actually migrates into 32l(s)). a substance is excluded from the to require appropriate action to restrict food! .definition oj "food additive" If its use the use of vinyl chloride polymers. Bie The Commissioner also concludes that was sanctioned bv H)A Prior to Commissioner concludees thflat the use of there Is no reasonable expectation of temher 6. 1358. A number of uses of Viny, vinyl chloride polymers should be 1proh..l.b..l.t.e..d....w...here " ' a reasonable ex- migration of vinyl chloride from jar and b_o_tt_le r_a_n__11_n_erns anTd ca_s_ke_ts. Pol.yvinyl consequently arc "prior sanctioned," pectatlon of Sation of vinyl ililQrtde cap liners and gaskets, whicIt Subsequent to the enactment of the chloride laty road. ,. , _ _ nlaavvee aalllmost completely replaced tne rub Food Additives Amendment of 1258, FDA This conclusion Is consistent with, the berf aandd, cork materials formerly used, are Issued letters stating that polyvinyl chlo requirements of the act for all uses of pofjjtywoo jmajor types. Some are applied in ride resin was generally recognized as er prior- liquid form as a ring around the part of sate In specific compositions, including the cap in contact with the container, rigid and semirigid articles Intended to and others consist of a circular disc cut contact foods. These letters were based generally recognized as sate. from film and inserted so as to cover on the premise that the uses and data The Commissioner interprets section completely the inside surface of the cap. cited in the Lehman articles for film and 402(a) of the act, which prohibits use of A majority of those types are plasti^ols coatings could be interpreted to extend food-contact articles that may render which are applied as liquid and are made to other food-contact articles containing food Injurious to health, as requiring a from paste resins containing finely polyvinyl chloride. showing of both possible migration and ground U micron or less) polyvinyl In addition, since 1953, a variety of possible harm. The Commissioner con chloride and plasticizer. These plaetisols uses of Vinyl chloride polymers m food- cludes that the criterion of migration in contain about 100 parts polyvinyl chlo contact articles have been approved_bv section 201 (s) of the act Is appropriately ride and 60 parts plasticizer. Other gas the Issuance of food additive regulations used in applying section 402(a) of the kets are made by combining these plasti- in 21 CFR Part 1217 Subparts D-&a]nd' E" act. Consequently, a poisonous or delete sols with other polyvinyl chloride resins. 121.1179 Coatings on Iresh citrus rious component of a prior-sanctioned In addition to removal of residual vinyl fruit; } 121.2507 Cellophane: i 151.2514 food-contact article comes within the chloride in the plasticizing process, addi Fcslnous and polymeric coatings; J 121.- terms of section 402(a) of the act If it tional vinyl chloride is thought to be re Adhesives; S 121~55?1 vtnui chin- may reasonably be expected to become moved when the plastlsol ts heated to ap ' 121.2524 a component of food. Since the carcino proximately 350' F for 5 to 8 minutes 5 121.- genic potential of vinyl chloride upon in during application. The small potential vansr- gestion is already sufficiently well docu for residual vinyl chloride that exists nooadrds: in1 1c2o1n.t2a5c4t3^ WitPhaacykuaegoinugs amnadterials mented to warrant a determination that it may, if present, render food injurious after such processing, together with the fact that a gasket has only limited con fgoWr usfeodoudsri:ng1 1th2e1.3ir5r4a5diaTtieoAn rouflsv;reTniaSclk.-- to health, the only prior-sanctioned uses of vinyl chloride polymers that may con tact with food, leads to the conclusion that there is no reasonable expectation 2550 Closures with scaling gaskets for tinue to be authorized are those where of migration of vinyl chloride into food. food containers: 8 121.2569 Resinous there is no reasonable expectation of Similarly, no migration may be expected meric coatings for voluolefln migration. from cap inserts cut from thin plasticized _ M2il1..25571 components ofvaHer Because of the likelihood that vinyl film, for the reasons previously discussed. JUL paperboard in contact with dry food: chloride is a carcinogen when Ingested, Moreover, In the case of all cap liners 5__1_2_1_._2_5_9_1 Semirigid aad rigid acruiic for uses approved by food additive regu there will be only slight contact with and modified acrufic plastics: i 120588 lations a fair evaluation of the data be food. Considering these factors, the Com Vini/l chloride-lauri/l vim ether. fore the Commissioner fails to establish missioner concludes that there Is no rea- voturners; 11212609 Vinyl chloridF- their safety wherever there Is reasonable foB&Me expectation of migration oHvffiyl ethvlene copolymers; } 121.2623 Vinyf expectation that vinyl chloride will mi chloride from can liners. chiortae-hexene-i copolumcrs: i 121.2631 grate from the polymers into food. There Can coatings containing polyvinyl Microporous polymeric filters. fore, the only uses that may continue to chloride are primarily used on the inside safely of a substance used In food- be approved we thosa whore there-LLno of beer and soft drink cans and, to a contatl Irtjfe ^ reasdriaoie expectation of such migration. much lesser extent, inside food cans. any tUimmeT. Uspe ofl a prior-sanctiloned sui b- For uses of vinyl chloride polymers Most of the polyvinyl chloride used for fitaace am be .prohibited aL limited mav be esttaabblliisshhe*d for its safeuse un; that have been generally recognized aa can coatings is produced by the solution safe, when there Is a reasonable expecta polymerization process which produces section, 402(a) of tion that vinyl chloride will migrate into polyvinyl chloride with the lowest resi l&llSMa. mmtssioner aetermWw food, the evidence of potential carcino dual vinyl chloride content. After con that W mayt be min. jurious to h- ealth genicity upon Ingestion requires the con version of the resin into can coatings, no For a substance uised pursuant to a foooad clusion that general recognition of their residual vinyl chloride has been reported, additive regulation, under section 409 of safety does not exist. Thus, as In the uses presumably because the thinness of the the act (21 U.S.C. 348) approval must that are prior-sanctioned or approved by applied film and the baking it has re be revoked when a fair evaluation of the food additive regulation, the only per ceived. at above 300* F, have caused the data before the Commc sioner fails to es missible uses are those where there lsnp removal of the residual vinyl chloride. In tablish that the substance is safe under reasonable expectation of such mlEi ~ such a case, it can be concluded that Its conditions of use. In the case of a sub tinn. there is no reasonable expectation of stance that is neither pnor-sanctloned In considering whether particular migration of vinyl chloride into food. nor the subject of a food additive regula food-contact articles raise a reasonable Polyvinyl chloride flexible tubing, tion, use may continue only as long os expectation of migration of vinyl chloride ranging in internal diameter from 2 toT die substance is geneially recognized as Into food, the Commissioner has reached thousandths of an In_c__h__to___3 ttoo 44 inclie~ safe. several tentative conclusions upon which )lv pPllaassticc)iz; ed. As Previously dis The Commissioner has reviewed the this proposal is based. cussed, i| t is thought that plasticization uses of vinyl chloride polymers In light of The Commissioner conclude^ tfyu there reduces residual vinyl chloride content (1) the available data concerning the is no reasonable expectation of migration to the point wlierci there la no reasonable safety of vinyl chloride and (2) the like- pf vinyl chloride from thin plastlcizecT expectation that any will migrate into FEDERAL REGISTER, VOl. 40, NO. 171--.WEDNESDAY, SEPTEMBER 3, 1975 VEV-144054 A <103:; 1 PROPOSED RULES afiKx;. Xu ariciiuou. flexible tubing Is gen ers are applied directly to the fruit. of phi tici'/iitioii, method a'. , y.-,mi-:i.'n- erally used In applications where food aerefore. the Commissioner concludes tion used, vinyl chkuide content of the contacts the tubing only briefly and the that the available data do not demon "compound" used, heat applied during temperature Of the lood Is often low. strate that this use is safe. " proce-sinc, and similar criteria. Com S circumstances tether reduce the Because copolymers of vinyl chloride ment.'. lioultl include all i:\ailnble data ilifcv of any migration of vinyl might also be expected to contain resid and information tint helu to identity chloride. ual vinyl chloride capable of migrating paiticular iippMr.it ions and , : ifimnnhs Tex tryIs incorporating vinyl chloride to food, this proposal applies to vlnvl that assure no rcisnnnble expectation of polymers in accordance with 9 121.2545 .chloride copolymers, as well as to the migration. The Commissioner advices (21 CPR 121.2545) arc nonwoven sheets homopolymer, polyvinyl chloride. " that at the time the final regulations arc prepared from natural or synthetic fibers In accordance with these conclusions, Issued, based on such datn and informa and bonded with flbrvls. The fibryls con the proposed regulations take the follow tion. it may be npptuprialc to restrict O'- sist of vinyl chloride-vinyl acetate copoly ing approach: eliminate uses that me here propo'-rd to mer resin that is prepared by solution 1. Prior-sanctioned uses of vinyl chlo- be continued. polymerization, a process that lias been ri3e~Tiomopolymers and copolymers~as Determining whether any food pa< k- shown to result In less than 1 ppm vinyl coatings. gaskets, can liners, flexible tub aging component, such as vinyl chloride. chloride In the resin. The fibryls nre ing. and niasticized turns would be idefi^ Is reasonably expected to become a com formed by precipitating a solvent solu tifled in reeuiations permitting UteTr ponent of food necessarily involves fine tion of the copolymer In water and then continued use This proposal identifies all Judgment, for which precise standards wasliing the precipitate until all solvent such prior sanctions knowq to the Cfi^ cannot be articulated. If there Is no de is removed. The flbryl is commingled with i mtssioner, Persons aware of oyer prior tectable vinyl chloride In a food-contoet fibers prepared from polyethylene .sanctioned uses should submit proof of article, and there are no detectable terephthalate resins to facilitate sheet the sanctions during the period for com? extractives of vinyl chloride from the formation and subsequently heat cured -ment on this proposal, article into food-simulating solvents, to fuse the flbryl and effect bonding. 2. Prior-sanctioned uses of vlnvl chlo and there is a sound theoretical basis These procedures for manufacturing ride homopolymers and copolymers InI for predicting no migration below the textryls should result in a reduction of I no detectable level, e g., the article is plas level of the residual vinyl chloride to the ticized or contact with food is slight, the point that the Commissioner concludes that their use would not reasonably be amend 9 i21.l06 (21 CFR I2l7l06) o/toe Commissioner concludes that there is no reasonable expectation of migration. expected to result in vinyl chloride be regulations to so provide. Once these Where vinyl chloride Is detected at a coming a component of food. proposed regulations become final, vinyl very low level in the food-contact article, Microporous polymeric filters used in chloride polymers could be used in semi it may nevertheless be possible to con accordance with 4 121.2631 (21 CPU 121.- rigid and rigid applications only after clude that there is no reasonable expec 2631) are based on polyvinyl chloride approval of a food additive petition, BUb- tation of migration into food based on resins produced by solution polymeriza mitted pursuant to 9121.51 (21 CFR theoretical considerations peculiar to tion. which, as noted above, results in 121.51) of the regulations. In addition to the particular product and use. The low levels of residual vinyl chloride. The the other required information for a food Commissioner advises that the detection filters are prepared by adding silicon additive petition, data would be neces of vinyl chloride extractives in food- dioxide to a solvent solution of the res sary to demonstrate that there is no simulating solvents under testing condi ins, resulting in "opening" of the resins reasonable expectation that vinyl chlo tions appropriate for food-contact arti and further loss of residual vinyl chlo ride will become a component of food. cles Indicates that the residual vinyl ride. The filter is formed by extrusion 3, Uses of vinyl chloride homopolymers chloride In the article may reasonably be and calendering followed by a hot syater and copolymers ns coatings, gaskets, cap expected to migrate Into food. The Com wash to remove solvent. Each of these liners, flexible tubing, and Plasticized missioner concludes that testing con steps, together with the preuse treat Jims tliat are not prior-sapeflonea, 553* ducted with food-simulating solvents Is ment required by 9 121.2631, would re an appropriate method for ascertaining sult In a reduction of any remaining the likelihood of migration from a food- vinyl chloride to the point that the Com contact article to food. Because of ana missioner concludes that the use of mi lytical difficulties, food often cannot be croporous polymeric filters would not reliably tested for evidence of migration. reasonably be expected to result in vinyl such uses as generally recognized as safe For tills reason, food-simulating solvents chloride becoming a component of food. may be submitted pursuant to 4 121.40 have long been used both by industry Adhesives containing vlnvl chloride (21 CFR 121.40). aaBuiicau 4. Food additive regulations pcrmltt and FDA to test food-contact articles. 'he Commissioner is aware that the use of adi77TWimrfi.fiTOiaiBaEnp-gra ym- R M-ft EftttL- DOfiowev|er.ffie data Iqdicfrto that rlijld OJlQ ggg intend*__ 3. udlng bot- .Ufa. .Miste'Ir ppefe. pna pipage*- ccpt as noted below tor water pipe) may .transmit vinyl chloride to ye food they Sintact, -tneretore, toe commissioner nda that these uses can no longer be permitted for contact with food. The use of vinyl chloride polymers as coatings for fresh citrus fruits, which Is permitted by 9 121.1179 (21 CFR 121.- 1179), presents the possibility of inges tion of vinyl chloride because the poly- in. water pipe as discu 5. Poo1daadddilitive reguiaUons specifically providing for the use of adjuvants In the production of food-contact articles con taining vinyl chloride polymers would be amended to be consistent with the pro posed restrictions on the use of vinyl chloride polymers. At the time final regulations are issued, it may be necessary to define the classes of permitted polyvinyl chloride foodcontact articles with greater particular ity. Thus, based on available data and on Information concerning theoretical pros pects of migration of vinyl chloride, spec ifications for permitted articles might be established in terms of thickness, degree classes of food-contact articles may exist for which It can be concluded that there la no reasonable expectation that vinyl chloride would migrate into food. Com ments on tills proposal suggesting tliat such articles do exist should Include data, analytical methodology used, and a theoretical analysis of the expectation of migration. In the case of polyvinyl chloride po table water pipe, the Commissioner con cludes that the data available at this time indicate that vinyl chloride may not reasonably be expected to be present in water drawn from a polyvinyl chloride water pipe system. Although data from the testing of polyvinyl chloride water pipe containing static water have shown FEDERAL REGISTER. VOL 40, NO 171--WEDNESDAY, SEFTEMSER 3. (975 VEV-144055 A PROPOSED RULES IOo-jj migration of vinyl chloride, no vinyl posed regulations, and has found that the paragraph (b)(3), and deleting the ref chloride has been detected in samples proposed action would not cause a major erence to paragraph (b)(3) from para of water drawn from operating poly Inflation impact as defined in OMB Cir graph (b)(4). vinyl chloride potable water pipe. It la cular A-107. Therefore, no Inflation im 2. By adding the following new sec likely that static testing does not rea pact statement Is required. At the time tion to Subpart E. to read as follows: sonably assess the likelihood of the additional announcements concerning 121.2009 Vinyl chloride polyntrr presence of vinyl chloride In water. It Is cosmetics, drugs, and medical devices are proposed that an interim period of time Issued, these conclusions will be reevalu resins. be provided for the continued use of ated. Data and Information concerning Polyvinyl chloride resins consist of polyvinyl chloride water pipe, pending environmental and Inflation Impact may basic resins produced by the polymeri development of data from tests appro be submitted as a comment on this pro zation of vinyl chloride. Polyvinyl chlo priate for the determination of the po posal. Copies of the FDA environmental ride basic resins have a maximum vola tential for the presence of vinyl chloride and Inflation impact assessments are on tility of not over 3 percent when heated in water drawn from a polyvinyl chloride file with the Rearing Clerk, Food and for 1 hour at 105* C, as determined by potabie water pipe system. Drug Administration. ASTM Method D 3030-72,' and an in Under the proposal, polyvinyl chloride A petition to ban the use of polyvinyl herent viscosity of not less than 0.35 as water pipe would be subject to the pro chloride in food packaging was received determined by ASTM Methods D 1243- visions of $ 121.4000 (21 CFR 121.4000). by the Commissioner on July 7,1975 from 66.' Vinyl chloride copolymer resins are concerning food additives approved on Public Citizen's Health Research Oroup. the polymers produced by the copolym- an interim basis. Within 60 days follow 2000 P St., NW, Washington, DC 20036, erlzatlon of vinyl chloride with other ing the effective date of a Anal regula as this proposal on the use of vinyl chlo monomeric substances. Vinyl chloride tion. an interested person would be re ride polymers was being prepared. Each homopolymers and copolymers may be quired to satisfy FDA that studies have of the petitioner's comments has been safely used as follows: been undertaken to determine whether considered in the drafting of this docu (a) Coatings, d) Polyvinyl chloride vinyl chloride may reasonably be ex ment. A letter will be sent to the peti for use as a can enamel. pected to be present in water drawn tioner responding to the petition. (2) vinyl chloride-vinyl acetate co from a system containing polyvinyl Copies of the reports and data referred polymer for use as a can enamel. chloride pipe. If no such commitment to above are on file at the office of the (3) Vinyl chlorlde-butadlene-aerylo- were made, or adequate and appropriate Hearing Clerk, Food and Drug Adminis nitrile copolymer for use as a component studies were not undertaken, the regu tration. Rm. 4-65, 5600 Fishers Lane, of conveyer belts Intended for use with lation permitting continued use of poly Rockville. MD 20852. fresh fruits, vegetables, and fish, and vinyl chloride water pipe would be Published elsewhere In this issue of as a component of coatings of paper and revoked. This announcement provides 60 days the Federal Register Is a notice with drawing a proposal to add } 121.2009 (21 paperboard In contact with meat and . lard. for public comment, after which time CFR 121.2009) and terminating the rule (4) Vinyl chloride-vlnylldene chloride the comments will be reviewed and Anal making proceeding on the use of poly copolymer for use as a liner, i.e., coating, regulations issued. The Commissioner vinyl chloride resin In articles for use for steel pipe. proposes that the regulations become In contact with alcoholic foods, which (b) Plasticised films. (1) Polyvinyl effective 30 days after their promulga tion as Anal regulations. No recall of was begun on May 17,1973 <38 FR 12931). Therefore, under the Federal Food, chloride for use In plasticized film in contact with food. affected articles Is now anticipated to be Drug, and Cosmetic Act (secs. 201 <s>, 402, 2) Vinyl chloride-butadJene-acrylo- necessary. The Commissioner concludes 409. 701. 52 Stat. 1042, 1046-1047 as nitrlle copolymer for use in plasticized that the hazard to the public health is amended. 1049, 1055 (21 U.S.C. 321 <s). film in contact with oleomargarine. not so Immediate as to warrant issuance 342, 348, 371)) and under authority dele (3) Vinyl chloride-vinylldene chloride of these reguls' ms without opportunity gated to him <21 CFR 2.120). the Com copolymer for use in plasticized film in for public com:..ent or to require recall missioner proposes to amend Part 121, as contact with food. and destruction of foods already pack follows: (4) Vinyl chloride-vinyl acetate co- aged. The continued use of installed equipment having food-contact surfaces 1. In 1121.106 by adding new para graph (e) (4) as follows: polvmer for use in plasticized film in con tact with food. composed of vinyl chloride polymers 3. In 9 121.2507, by amending para would be permitted; any residua] vinyl chloride Is likely to have dissipated to S 121.106 Substances prohibited from tue in human food. graph (c) by revising the entry In the list of substances for "polyvinyl chlo the atmosphere during the period of * ft ride" to read as follows: service. These proposed regulations deal only with vinyl chloride contamination of food. The Commissioner plans to issue (e) * * (4) Vinyl chloride. (1) Vinyl chloride has the molecular formula CiHjCI. it is a synthetic chemical not found In natural g 121.2507 Cellophane. * I (C) * additional announcements in the near products and has been used In the pro * * *a future concerning cosmetics, drugs, and duction of polymeric substances that Limitations * * * medical devices. Also, the proposed regu may contact food. * lations would not Immediately affect the (11) Food containing any added or de Polyvinyl chloride.. As the basic polymer status of vmyl chloride polymers used tectable level of vinyl chloride Is deemed for use only in coftt- In food-contact articles In the household, to be adulterated In violation of the act. lg9. food service establishments, and food dis (ill) The use in food-contact articles * t * pensing equipment. Such articles are the subject of a notice published in the Fed eral Register of April 12, 1974 (39 FR 13285), and they will be evaluated In ac cordance with the terms of that notice. of vinyl chloride homopolymers and co polymers Is prohibited, except that such use Is not prohibited: <a) In coatings, gaskets, cap liners, flexible tubing, and plasticized films if 4. In 1121.2511. by amending para graph <b> by revising the listing entries for "dicyclohexyl phthalate" and "di phenyl phthalate" to read as follows: The Commissioner lias carefully con such use Is otherwise in accordance with 121.2511 Plasiici/cri in pel;, meric sidered the environmental effects of the the requirements of the act and this tih-taneca. proposed regulations and, because the chapter; or * proposed action would not significantly <t>) If specifically permitted in this affect the quality of the human environ part. (b) ment, has concluded that an environ mental Impact statement Is not required. The Comn.:.'loner has also carefully con sidered the Inflation impact of the pro- 121.1179 [Amcmh'il] la. In S 121.1179 Coating) on fresh citrus fruit by deleting and reserving Copies may bs obtained from: American Society for Testing and Materials, 1913 Race St., Philadelphia, PA 19103. t FEDERAL REGISTER, VOl. 40. NO. 171--WEDNESDAY, SEPTEMBER 3, 1975 U8M 44056 4 ATTACHMENT B VEV-H4057 14464 Federal Resister / Vol. 47, No. 64 / Friday, April 2, 1982 / Proposed Rules * DEPARTMENT OF HEALTH AND HUMAN SERVICES Food and Drug Administration 21 CFR Chapter I [Docket No. SIN-0281] Policy for Regulating Carcinogenic Chemicals In Food and Color Additives; Advance Notice of Proposed Rulemaking aorncy; Food and Drug Administration. ACTION; Advance notice of proposed rulemaking. summary: The Food and Drug AdministrationtkUA) is considering proposing a new policy tor use in assurinathe safeuso ot lood additives undcoloradditives that contalrTmmor amounts ol carcinogenic cnemicais. mis policy would replace the current caseby-case approach to regulating such additives with an approach that applies consistent standards. This advance notice describes the scientific and reje'^tory problems confronting the agency with respect to such additives and explains the need to consider a change in FDA's present approach. This advance notice also discusses the alternatives that the agency is considering. FDA is publishing this advance notice so that the agency will have the benefit of a broad range of views early in the rulemaking process, and so that interested persons will have an opportunity to express their opinions. Detailed regulations for implementing any policy that the agency may adopt will be proposed in a future issue of the Federal Register after consideration of comments received in response to this advance notice. Elsewhere in this issue of the Federal Register, the agency is publishing a final rule listing D4C Green No. 6 for use in externally applied drugs and cosmetics. That rule further explains and illuslratea how FDA has applied the approach set forth in this advance notice of proposed rulemaking. This document also discusses alternative theories for implementing the approach that the agency applied in the decision on D&C Green No. 6. OATH Comments by July 1,1902. ADDRESS: Written comments to the Dockets Management Branch (HKA305), Food and Drug Administration. Rm. 4-62. 5600 Fishers Lane. Rockville. MD 20857. FOR FURTHER INFORMATION CONTACT: For information relating to the policy under consideration and regulations to be proposed: Terry C. Troxell, Bureau of The Food Additives Amendment Foods (HFF-334), Food and Drug defines a food additive as: Administration. 200 C St. SW.. Washington. DC 20204. 202-472-5690. For information relating to risk * * * any substance the intended use of whit.h results or may reasonably be expected to result, directly or indirectly, in its assessment procedures: Alan M. Rulls, becoming a component nr otherwise affecting Bureau of Foods (HFF-334), Food and Drug Administration, 200 C St. SW., Washington. DC 20204, 202-472-5676. .SUPPLEMENTARV INFORMATION: the characteristics of any fund '`n-i-auing any Substance intended fnr use in producing, manufacturing, pocking, processing, preparing, tieating, pickaging, transporting, or holding food: and including any source of I. Background radiation intended for any such use) * * * 21 U.S.C 321(a). Like all chemicals, no food additive or color additive can be produced Thus, the amendment covers both absolutelvpurc. Each additive, even a direct additives Substances intended highly punned one. consists of an array for direct use in foodsl and indirect of chemical entities, both functional and additives (substances that tnrougn use nonfunctional. For example, a color in food packaging or for other tood- additive may contain ona primary and a contact purposes might become or might number of secondary colors, all of which reasonably be expected to become are somewhat functional In the final Mmgonents^offood^ additive, A polvmer resin used in food The legislative history makes it clear g may contain a continuum ol that Congress intended the amendment Sa from mo'npmer*an5*55nerTo to protect consumers from unsafe food gcular weight polymers, only additives while at the same time some of which are functional. Finalplye,ramllitting manufacturers to use food jJ}2nicalsubstancestJnciudiiigJhosaedditives at safe levels (H.R. Report No. used as additives, contain numerous 2284. B5th Cong.. 2d Sess. (1958)). This jmourities such as residual reactants, intent is reflected in the standard for intermediates, manufacturing aids, and approval called the "general safety products of side reactions and clause," that Congress inserted into the fogradatiom law: in the early 1050's Congress became No such regulation shall issue if a fair concerned about the safety of the evaluation of the data before the Secret,iry- chemicals being used in increasing amounts in foods. Under the adulteration provisions of the lawt In order to prevent the use of such chemicals in food. FDA waa required to (A) fails to establish that the proposed use of the food additive, under the conditions of use to be specified in the regulation, will be safe * * * U.S.C 348(c)(3)(A). prove in court that a food bore or The safety of a food additive must be contained a poisonous or deleterious demonstrated by properly conducted substance which may render it injurious and evaluated scientific tests, but the to health (21 U.S.C. 342(a)(1)). This additive need not be shown to be process was cumbersome and time- absolutely safe. Congress recognized the consuming. More important, it did not impossibility of proving the safety of an adequately protect the consumer, additive beyond any doubt and under because FDA had the burden of proving any conditions of use. Reflecting the that the substance used in food had viewpoint of the various scientists who - some potential for causing injury. testified in support of a reasonable Manufacturers also disliked the system standard for the safety of an additive, because it prohibited the use of any the House of Representatives Report on amount of a chemical substance it that the amendment states: chemical had the potential for causing Injury. Accordingly, In response to these The concept of safety used In this legislation involves the question of whether a substance is hazardous to the health of man mounting concerns from consumers and or animal. Safety requires proof of a manufacturers alike. Congress passed the Food Additives Amendment to the Federal Food, Drug, and Cosmetic Act (the act) in 1958. (21 U.S.C. 321(a) and 348.) That amendment set up a premarketing safety clearance system requiring that a food additive be shown reasonable certainty that no harm will result from the proposed use of an additive. It does not--end cannot--require proof beyond any possible doubt that no harm will result under any conceivable circumstance * * * H.R. Report No. 2284,85th Cong., 2d Sess. 1. 1958. to be safe for its intended use by the FDA regulations governing the safety sponsor and be approved by FDA before assessment of food additives have it is used in food. The legislation also incorporated that language. Under 21 established safety standards to apply to CFR 170.3(1), safety of a food additive the evaluation of the additive. means that there is a reasonable 1 Federal Register / Vol. 47, No. 64 / Friday. April 2, 1982 / Proposed Rules 11465 ortainty in the minds of competent rantists that the substance is not J under the intended conditions 4 use. Because of particular concerns about oncer. Congress added_the anticancer danse, or Delaney Clause, to the Food Additives Amendment. This Clause establishes a flat rule "(t|hat no additive be deemed be safe if it is found to induce cancer when ingested by man or animal or if it is found, after tests which ere appropriate Tor the evaluation of the safety of food additives, to induce oncer in man or animal * * *" (21 USC 348(c)(3)). Although the Department of Health, Education, and Welfare (HEW) (now the Department of Health and Human Services) did not object to the inclusion of the Delaney Clause in the act the Assistant Secretary for Health made it dear that, in the view of HEW and FDA, "the (food additives amendment) reads and meant the samp with or without the inclusion of the (Delaney) clause * * *." H.R. Rept. No. 2422. 85th Cong., 2d Sess., pp. 10-11. The Department held this view because of the prevalent scientific conviction in 1958 that the state of the art would not permit scientists to establish a tolerance for a carcinogen. Thus, it was impossible to determine reliably a level of exposure to a carcinogen below which one can be assured that three is no significant increase in the risk of developing cancer. In other words, scientists lacked the technology to assess adequately the risk presented by a particular chemical. (See Food Additives; Bills to Amend the Federal Food, Drug, and Cosmetic Act with Respect to Chemical Additives in Food; Hearings Before the Subcomm. of the House Comm, on Interstate and Foreign Commerce, 85th Cong., 1st Seas., 369,1957 (sta'ement of W.C. Heuper, M.D.).) The Color Additive Amendments of 1960 (21 U.S.C. 321(t) and 376) establish standards for the safety of a color additive that are virtually identical to those set forth in the food additive provisions except that they contain a slightly different version of the Delaney Clause for neningested color additives (21 U.S.C. 37C(b);5)(B)). In addition, the Color Additive Amendments provide for the premarket approval of color additives in drugs, riewci s. and cosmetics as well as food through a mechanism Called listing. For a discussion of the Color Additive Amendments, see the final rale publi-.hed elsewhere in this issue of the Federal Register listing D&C Given No. 6 for use in externally applied drugs and cosmetics. II. Regulatory History During the two decades that FDA has administered the food and color additive provisions, the agency has. with a few exceptions, interperted the food and Color Additive Amendments to ban the use of any additive that was found to contain or was suspected of containing minor amounts of carcinogenic chemicals, even If the additive as a whole, had not been found to cause ` cancer (noncarcinogenic additives). For example, FDA terminated the provisional listings of carbon black (41 FR 41857; September 23,1976) and graphite (42 FR 00734; November 29, 1977] because the agency suspected that these colors could contain polynuclear aromatic hydrocarbons (PNA's). some of which are carcinogenic. The agency removed Ext D&C Yellow No. 1 from the provisional list (42 FR 02478; December 13.1977] because of the possibility that it might contain as impurities 4- aminobiphenyl and benzidine, both of which have been shown to be carcinogenic In humans. FDA terminated the provisional listing of D&C Red Noe. 10,11,12. and 13 because the agency suspected that they might contain low levels of -naphthylamine, a known human carcinogen (42 FR 62475; December 13,1977). The anenev has also proposed to i t itlffUliTku l l i`)s;*)vt'iuwDnPonrtid* e iu'/ril in contact with food (40 fW* hiTl)iTTirnTJ f3,1 because of e monomer, a known-human carcinoBen. could migrate tnjmali amounts into food froraFVC The agency has approved the use of food and color additives that contain or may contain minor amounts of lead and arsenic, proven carcinogens. Those chemicals are ubiquitous environmental contaminants, routinely found in food and water as well as in chemicals used to manufacture food end color addities. Their pervasiveness is such that no food or color additive can realistically be manufactured without the possibility of some lead or arsenic contamination. Thus, because the alternative would be to prohibit the use of all additives, the agency has minimized the amount of the chemicals and thus the risk attendant to the use of such additives by setting specifications for lead and arsenic in many food and color additive regulations. FDA does not regard these exceptional situations as constituting a true departure from its traditional regulatory approach described in the examples listed above. In sum. the agency has generally concluded that a safe level could not be established for a carcinogenic chemical in a food or color additive and has consistently taken regulatory action against additives that contain carcinogenic chemicals. In an early administrative interpretation shortly after the enactment of the Color Additive Amendments, the agency construed the Delaney Clause as being triggered only if tests of the color additive as a whole revealed that the color caused cancer. (Discussion of Sec. 8.36, Digest of Objections to |anuary 24,1961. proposed color additive regulations.) See the final rule published elsewhere in this issue of the Federal Register listing D&C Green No. 6 for use in externally applied drugs and cosmetics. It is not clear, however, to what extent, if any, this interpretation wae applied In subsequent agency decisions on specific color additives. IH. Context in which Regulatory Alternatives are Presented Recent^clentifi^andjegal developmenTTnaveconyTricedFDA that 5 may nowbepeceisary and" ftRBIBBQftVT WfihtljKr how It i Si t contains i figaMwmtaaj carcinogenic- Over the past 20 years, there have been rapid developments in analytical capabilities that make it possible to decrease by orders of magnitude the levels at which the components of a substance such as a food additive or color additive are detectable and identifiable. Many chemicals now can be identified and quantified at levels around one part per billion. Coupled with this development has been a large increase in the number of substance! that have been studied for carcinogenicity in animal bioassays. For example, Tomatis has reported that 826 chemical substances were under test for cancer throughout the world in 1975 (Ref. 1). Many of these bioassays have resulted in positive findings for carcinogenesis.1 The Occupational Safety and Health Administration has reported (Ref. 2) that about 17 percent of 7,000 test chemicals on a Public Health Service list are tumorigenic. Criesmer and Cucto have reported that 52 percent of the chemicals that have been studied in the National Cancer Institute (NCI) 1 ihe percentage of positive bioassay results in the reports cited is high partly because the chi micitli frCicUcd for evolution .mo t :ho?e tiut u.-e suppeUed of bi.ing carcinogenic. Scientists select these cnttnicals on the basis of available information, such as molocuUr structure'uctwity relationships, genotoxicity studies, and cpultfMiulng cal evidence. VEV-144059 ,r : - *c , V' - ^? -i Federal Register / Vol. 47. No. 64 / Friday. April 2. 1982 / Proposed Ruk-s t bioassay program are carcinogenic (Ref. 3). A National Toxicology Program report indicated that 41 percent of 95 tests completed showed that the tested substances are carcinogenic (Ref. 4). As the number of chemicals that are found to cause cancer in animals has grown, and as scientists' ability to detect the components of a substance has become more acute, the chances that a food additive or color additive will be found to contain a carcinogenic chemical entity increase. It iB the agency's frank expectation that a growing number of additives will be found to contain a carcinogenic chemical in future years. If FDA continues to implement the regulatory approach that it has followed in recent years, it will be forced to refuse to approve or to terminate the approval of the use of each of these additives, even though they themselves may be safe. However, the agency believes that there are alternatives to its current policy that will adequately protect the public health. Not all of the additives that have been or will be found to contain carcinogenic chemicals will themselves be shown to induce cancer in appropriate tests (e.g.. D&C Green No. 61. the aaenev believes that a tmction can be drawn oetween additives that contain a carcinogenic cnemical Eu'nhaTVave not themselves been shown to be carcinogenic. Two recent developments support the agency's belief that such a distinction Is appropriate. One development was the 1979 decision by the United States Court of Appeals for the District of Columbia in Monsanto v. Kennedy, 613 F. 2d 947 (D C. Cir. 1979). In discussing whether a substance that migrates into food is a food additive, that court expressed the view that there is "administrative discretion, inherent in the statutory scheme, to deal appropriately with tfe minimis situations." Id. at 953.." hy discretion to dipissrrefgarpdtllfoliwiiu-llei vel mmigipralitpiofnfjInnjtoo fooqdTgof siMubSsgtaianpceasHiJna, indirect additives because the migration gtnaVarUcuiarja^injj_greMhlrftA public health concern, then me agency may also disregard, alter apprbprBfT tests, a carcinogenic cnemical tna easonable certainty of no harm Irom* 'chemical. SeconSTTRe agency is now confident it it possesses the capacity, through i use of extrapolation procedures, to less adequately the upper level of risk isented by the use of a ^carcinogenic additive that contains a 'cinogenic chemical. FDA has been uctant to use these procedures for regulatory decisionmaking. Many theoretical models have been developed to extrapolate from animal experimental data to the relatively low levels of possible human exposure, but they can vary widely in the risk values that they predict. Thus, knowledge of the true risk at relatively low exposures is elusive. Nevertheless, even though there is an inadequate scientific basis for confidence in the accuracy of predictions of actual risk by these procedures, there has been a growing recognition in the scientific community that by using certain conservative extrapolation models it is possible to estimate an upper limit of risk. (See e.g., "Chemical Compounds in Food* Producing Animals" (44 FR17070; March 20,1979) and FDA's decision on lead acetate (43 FR 72112: Ocober 31,1980 and 46 FR 15500; March 6,19B1).) The estimate of the risk may be exaggerated by these conservative extrapolation models. Because the estimated risk will not be understated, however, such risk assessment techniques can be used with confidence to determine whether, under the general safety clause, there is a reasonable certainty that no harm will result from the intended use of an additive. IV. Regulatory Approach A. Introduction In the decision on D&C Green No. 6 published elsewhere in this issue of the Federal Register, FDA is approving a color additive that has not been shown to be a carcinogen in appropriate tests, even though it contains a carcinogenic impurity. In this advance notice of proposed rulemaking, FDA is announcing its intent to formally adopt the principles on which that decision is based as the general policy of the agency. Therefore, FDA is soliciting comments on whether it should do so, or whether it should adopt soma other approach to the problem. To facilitate such comments, the remainder of this document will describe the general policy that FDA is considering. The policy consist of three elements: 1. Clarifying exactly what an "additive" is; 2. Interpreting the Delaney Clause to apply only when the additive itself has been shown to cause cancer; and 3. Using risk assessment as one of the tools for determining whether the additive is safe under the general safety clause. FDA believes that the act and its legislative history, as well as subsequent judicial and administrative interpretations, support such a genera] policy. B. Construction of the Term "Additive" Conceivably, each chemical in the complex mixture that constitutes a food additive covdd itself be considered to be a food additive. Each of these chemicals in some sense becomes a component of the Food of which the additive is a part. For ex,-triple, in the case of an indirect food additive used in food packaging or the like, any chemical impurity that migrates from the indirect additive into food could be considered to be an additive. In the Monsanto case, the agency argued that the food additive definition applied to the residual acrylonitrile monomer used to fabricate the final bottle, as well as to the bottle itself, because the monomer was intended for use and was used to manufacture the copolymer bottle. (Government's Brief, pp. 63,67.) However, the Monsanto courthgyjji^ tlmitatjrie^doesnotMtngelj^ CflSmlssIonerorT'ood and uniga to declare that each chemical in an fladibyitis itself an additive. Monsanto, id. at 954, 955. FDA has reexamined the issue of when a chemical should be considered to be a food or color additive, and has developed three alternative but not mutually exclusive interpretations to address the problem. Each of these interpretations is presented here. FDA has worked out the first of these, the constituents approach, more extensively than the others, but the agency wishes to emphasize that, at this time, it does not view any legal theory as being preeminent. Indeed, FDA may conclude that one. all. or some other combination of approaches may be appropriate. The overall risk assessment policy outlined in this notice, however, would be the same under each interpretaton. 1. The constituents approach. This approach dittintfuishca between the additive as a whole and Its consIHuents for the purpose of determining when the Delaney Clause is triggered. Using this approach, the food additive would be the substance that is actually intended for use in food or for food contact. All nonfunctonal chemicals present in that substance would be called the JJmjflU_2|had3lGve" Similarly, m "color additive" would mean only those substances intended for use as a dye, pigment, or other substance that is capable of imparting color. (See 21 U.S.C. 321(t).) Any other chemical present in a color additive but not intended for use as described above would be a constituent of the color additive. WVlti ffltilrff residual reactants, intermediates, i ' as products VEU-144060 Federal Register / Vol. 47, No. 64 / Friday, April 2, 1982 / Proposed Rules 14467 aftide reactions and adationjAconatituei^ ^oimejujamveasa^EoJe^wmB metroe conaiaerea toTTST""* ilnvi for regulatory purposes because i not Intended forma In food or tor TMI_W_i_if_:l_i\J_mW_*._*r5_ri_dlt_*i_)-l_!#_!*)]m!_jnt_'fi_iTT_t*_T1- nP-c-lEs--.J-i-inn' no one wants such a constituent to TiiirniT-n men n an ^Tirrapnip.mm i reaction* 'irr^m^w^^E^n^^a.Enjntocompletfonjand idatfon or decomposition could be w1r-"fT-m.incrtiTnnQnt-BeSiai "r~ rrvmr-mtMs-i-nz [alRaFway. ^iiji ii-r-i'nnnrnimtiiT^r! th~ey abuelIiaBfnBip7n1aip.tIieTrIT1'"i!T'! grnTJEfnr-i cannt ).|!1 IS. i Rdmautii In thHeecawte of a direct food additive, rite identity of the additive is anambiguous: it is the substance Secluding all its constituents that the food processor deliberately puts into the food. Similarly for color additives the additive is the dye. pigment, or other sabstance Including all its constituents dut the manufacturer deliberately incorporates into the food, drug, device, er cosmetic for the purpose of imparting tha intended color. Some examples may clarify how nrious substances in direct food dditives or color additives would be dsssified under the constituents approach. A substance intended for use (a the processing of any food would be considered a food additive even though that substance may essentially be removed from the food in subsequent processing steps. Thus, extraction solvents employed in food processing would continue to be food additives because they are intended for use in a food. Such products are distinguishable from the residual reaction products such as p-toluidine present in D&C Creen No. A because p-toluidine is not intended for use in the final color additive but only in the nianufacture of the additive. I'i'tfjilr-M! Mina materials are rrritrwimrTJismtrnifog} ^re complex and that often contain a jyjmowoMubs^dnceSjejcn^r^hichis intended tq have a specific physical or eghQifgl'gQgclt For example, one type of plastic'food wrap contains slip agents <nd antioxidants that are blended with polyolefin resm and incorporated into the final polyolefin film. All three of these substances would be considered to be additives, because they are all intended for use in the final food-contact preparation. Conversely, a chemical used in the manufacture of the polyolefin resin end remaining In the additive as an impurity resulting from an incomplete chemical reaction would be treated as a constituent J2m2Q8til]lfiBtUBB8IGhJllSSLil|tR detectlontechnin|ies"AlthouBh it generally is possible to detect or predict migration to food of at least one constituent from an additive used In a food-contact surface and although the number of migrating constituents which are oetectabla continually increases with advances in analytical methods, it is not currently possible to identify completely ell the constituents that become components of food as a result of the use of an indirect additive. The detection, or prediction by sound scientific methods, of migration of one or more constituent* of an intended substance ia an indication that all of its ingredients may become components of food to some extent, however slight. The constituents approach also cognizes that the only practical way rtheaaencvto ensiyetnat CmatitMflnH Of the Indirect a which become comooinentsoffbod^g not exceed ante limits i Issue i raw etion that on[scribes m or thft ind asUsai edditiva as a whole, not merelyv flor the Two additives intentionally blended together for use In food, for food contact, or for imparting color, would remain additives; the physical mixing would not change one of the component additives into a constituent. The focus under the constituents approach, as in Monsanto, is on the safety of the additive aa a whole. FDA believe* that, although it differs in construct from the do minimis migration approach discussed below, the constituents approach is supported by the reasoning in Monsanto. 2. De minimis migration approach, A second approach, somewhat different from the constituents approach, is based directly on the Monsanto decision. The court, recognizing the difficulties posed by a strict literal interpretation of the statute, declared that "there is latitude inherent in the statutory scheme [of the Federal Food, Drug, and Cosmetic Act] to avoid literal application of the statutory definition of `food additive' in those de minimis situations that, in the informed judgment of the Commissioner, clearly present no public health or safety concern." Id. at 954. The Monsanto decision permits the CaoffllSSIflfliOfiSftifflBUsHSSSkiiJ that migrate at laYYltYskiniB [ogfl.iflb migr^attoioQnj2p{ij^migajtj2atiuiaufissiiisicaaLl.( tthhee lleeyveell iinn gQuueeasttion noses no threat to the public health, ynderflif <J99UM.il appears that the Commissioner may also take the position that the Delany Clause is not triggered by the presence of a carcinogenic chemical that occurs in a noncarcinogenic direct food additive or color additive, if the agency can conclude that there is a reasonable certainty of no harm from the presence of the chemical. Under this narrow reading of Monsanto, called here for convenience the de minimis migration approach, FDA would set a fixed upper limit on the absolute amount of a carcinogenic chemical that could be present.The existence of "reasonable certaintly of no harm" from the presence of that chemical is predicated on the fact that only a minute amount of it is present. That is. the risk from the de minimis migration of carcinogenic chemicals into food from an indirect additive (or the de minimis presence of such chemicals in a direct additive] is trivial because the amount of the chemical is so low, and the resulting potential human exposure to it is thus trivial. A policy based on the de minimis migration approach as set forth here would be narrower than one based on the constituents approach, ft would not permit the agency to approve the marketing of additives containing carcinogenic chemicals that are present in food in more than trivial amounts, even though the risk from the use of the additive falls below an acceptable upper limit as determined by an appropriate risk assessment evaluation. If the fixed upper limit of exposure to a compound (based on its concentration in a particular additive and its prevalence in other additives) is set relatively low, as Monsanto suggests it should be. then this policy would permit the continued use of very few additives with carcinogenic constituents and would result in few regulatory decisions that would differ from those that could be reached under the present policy. 3. Sensitivity of the method (SOM) approach. FDA believes that an approach analogous to that proposed for use in regulating the residues of carcinogenic drugs administered to food-producing animals may also justify a policy for regulating carcinogenic chemicals present in minor amounts in food additives and color additives that are not themselves carcinogenic. m 'id o u jo a i K m Aw n ^V. *44061 14468 Federal Register / Vol. 47, No. 64 / Friday, April 2, 1982 / Proposed Rules A proviso in the Delaney Clause for chemical compounds administered to food-producing animals (21 U.S.C. 340(c)(3)(A), 360b(d)(1)(H), 376(b)(5)(B)) allows the use in such animals of a chemical that is a carcinogen, provided that no residue of the chemical is found in edible portions by methods prescribed or approved by FDA. If a compound for use in food-producing animals is a carcinogen, the agency has proposed procedures referred to as the SOM (see 44 FR17070: March 20,1979) governing the collection of data showing whether the proviso has been complied with; i.e., capable of showing what residues may safely be allowed to go undetected. Applied to food additives and color additives, a policy based on the SOM ' approach would permit the Commissioner to decline to classify as an additive a carcinogenic chemical thought to be present in the additive, but undetectable by a specified method, as long awthe risk of cancer posed by that concentration of the carcinogen fell below a certain "acceptable" or "insignificant" level. That is. the detection of a carcinogenic chemical becomes the ultimate inquiry. If the chemical cannot be detected and if its hypothetical presence below the detection level does not exceed an acceptable risk of cancer, the chemical need not be regulated as an additive. The SOM approach is the narrowest of the three approaches described and would result in the least change from current policy.. C. Interpretation of the Delaney Clause The Delaney Clause requires the disapproval of any food additive that has been shown to be a carcinogen in appropriate testing. It states: That no additive shali be deemed to be safe if it (the additive) is found to induce cancer when ingested by man or animal, or if it [the additive) is found, after tests which are appropriate for the evaluation of the safety of food additives, to induce cancer in man or animat * * * 21 U.S.C. 3411(c)(3)(A). However, it does not state that an additive shall not be deemed safe if the additive "or any of the chemicals present in the additive" is found to induce cancer. Addition of the quoted language to 21 U.S.C. 348(c)(3)(A) would change the meaning of this provision. A natural reading of the language that does appear in the statute establishes that the Delaney Clause does not apply to a carcinogenic chemical in a food additive absent a finding, after appropriate tests, that the additive as a whole induces cancer. Similar reasoning would apply to the Delaney Clause in the color additive provisions of the act. (21 U.S.C. 376.) The legislative histories of the Food Additives Amendment and Color Additive Amendments are silent on this question. However, the agency's contemporaneous interpretation of the Color Additive Amendments supports the interpretation of the Delaney Clause set forth above, as explained in the preamble to the D&C Green No. 6 regulation. D. ProofofSafety Under the General Safety Clause As discussed in Section III above. FDA believes that risk assessment may be appropriately used to determine whether, under the general safety clause, there is a reasonable certainty that no harm will result from the proposed use of the additive. The risk assessment techniques under consideration are similar to those used by the agency and discussed in its evaluation of the safety of DftC Green No. 6 published elsewhere in this issue of the Federal Register. Interested persons are referred to that discussion for an example of how a safety determination using risk assessment would proceed. Because carcinogens vary widely in potency, and because human exposure to carcinogens usually occurs at levels several orders of magnitude below that at which compound-related human tumor incidence can be detected by scientific studies, it has been necessary for scientists to develop quantitative extrapolation procedures to estimate the levels of risk to humans of exposure to carcinogens at very low levels. Several quantitative risk assessment models such as the one-hit model (Ref. 5). the Crump modified multistage model (Ref. 8). and the linear proportional model (Ref. 7) are available to extrapolate animal bioassay data to human exposure. Any one of these models can be used to estimate the upper limits of potential risk created by exposure to the chemical. The agency has not determined which extrapolation model is the most appropriate one to use. end requests comments on this issue. After relevant animal data have been generated and analyzed, an appropriate risk assessment procedure would then calculate an acceptable maximum exposure level to the chemical based on a suitable risk standard. This level would be calculated by extrapolating from an upper confidence limit of bioassay data, using procedures that would include an evaluation of harmacodynamic data, including any uman studies available. I epidemiological evidence, and studies of the mechanism of action of the carcinogen. FDA is requesting comments on what risk standard would be appropriate for use in implementing the policy. Once FDA has determined a maximum acceptable level of exposure to a carcinogenic chemical, the agency would then consider whether the amount of actual exposure to the chemical from food additives and color additives is within the acceptable level. If necessary, the agency would develop specifications for such chemicals in individual additive regulations, taking into account total estimated exposures. Under the general safety clause, an additive must meet the agency's toxicological requirements for food additives and color additives. The risk assessment procedures discussed in this advance notice would not modify currently applied requirements under the general safety clause other than to provide a procedure for determining whether a noncarcinogenic additive that contains minor amounts of a proven carcinogenic chemical is safe. The risk assessment procedure would provide a method for estimating the levels of such chemicals that meet the general safety clause standard of safety. The upper limit of acceptable exposure would be determined by carcinogen potency and risk extrapolation by the best current scientific methods available. FDA believes that any risk assessment procedure should yield such low acceptable levels that nothing but minor levels of carcinogenic chemicals would be able to pass the screen. Risk assessment of carcinogens is currently not completely rigorous from a biological viewpoint, but it does provide, at minimum, a scientifically based means to evaluate the health concerns of a series of chemicals. Its weakness stems from the substantial assumptions which are employed. However many of these same assumptions are endemic to toxicology evaluations in general. In order to calculate a given risk level, a dose-response relationship is extrapolated to doses several orders of magnitude lower than the tested dose. Furthermore, the risk assessment assumes that humans will be no more sensitive to the carcinogen than were the most sensitive animals tested (Ref. fl). If the policy is implemented, it is FDA's cunent view that the risk assessment procedure should be conservative to compensate for the weakness in scientific rigor, should use a model that is likely to overestimate the UEV-144062 Federal Register / Vol. 47, No. 64 / Friday. April 2, 1982 / Proposed Rules 14469 r'k, and should contain a low risk lard. A believes that this general policy is a sensible, scientific means of limiting the circumstances in which it will be necessary for the government to act to ban the use of a food additive or color additive, without compromising the public health protection afforded by the act. This policy, if finally adopted, is intended to be implemented solely in those instances (e.g., in the case of D&C Creen No. 0) where data demonstrate that there is a reasonable certainty that no harm will result from the use of an additive that contains a carcinogenic chemical. E. Alternative Policies As this advance notice explains, FDA is considering formalizing an approach that will assure the safe use of food additives and color additives that contain minor amounts of carcinogenic chemicals, and requests comments on that approach. In considering how best to develop a policy on this issue, the agency has followed the line of reasoning set forth in this notice. The agency recognizes, however, that a number of different approaches might also deal with the problem, and requests t*'"t interested persons submit any native policies to FDA as ments. FDA is also aware that the three approaches (constituents, de minimis migration, and sensitivity of the method (SOM)) discussed above support regulatory decisions of varying breadth. In general, if used alone the constituents approach would be the broadest, and the SOM approach would be the narrowest. The agency requests comments on which of the three approaches or combination of approaches would be the most workable from a legal, scientific, and procedural perspective. Further. FDA is interested in receiving comments on any other approaches--either based on. or different from, the three that are discussed in this advance notice. Finally, FDA would like to receive comments on whether an across-theboard rule incorporating the suggested approaches should be issued or whether, instead, the agency should continue handling carcinogenic chemicals in food and color additives on a casc-by-case basis. An agency is free to choose between evolving an approach to a regulaiuty problem through a series of individual lions, on the one hand, and issuing ntral regulation to be applied in ecific cases, on the Other. Both methods have advantages and disadvantages, and the choice uf which to use is generally within the discretion of the agency. See Securities and Exchange Comm. v. Chenery Corp., 332 U. S. 194 (1947); NLRB v. Bell Aerospace Co.. 416 U S. 267, 294 (1974): Chisholm v. FCC. 538 F.2d 34ft 364-366, cert denied. 429 U.S. 890 (1976). In thia advance notice. FDA request* comments on how its discretion should be exercised in developing an approach to the handling of carcinogenic chemicals In food and color additives. The agency has a number of such Issues before it in addition to D&C Creen No. 8. on which there is a pressing need to take action. The agency doe* not view the existence of an ongoing rulemaking proceeding as posing any bar to specific regulatory actions. The agency emphasizes that it will continue to make regulatory decisions on the food end color additives that come before it in the ordinary course of business. In making such decisions, the agency may use one or more of the approaches discussed herein, as appropriate. V. Economic and Environmental Impact The risk assessment policy under consideration may have economic consequences for individual industries. In some cases, the policy should produce positive economic benefits that would result from preventing a ban of an additive containing a carcinogenic constituent. The agency ie not aware of any significant negative economic effects of this policy. If proposed regulations are to be issued after FDA considers the comments received on this advance notice, then FDA will be required by Executive Order 12291 and the Regulatory Flexibility Act (Pub. L 96-354) to access the economic consequences. However, the agency does not now possess the data that would permit e detailed assessment of the economic impact Therefore, the agency requests that interested persons submit information that will aid the agency in making an economic impact assessment. FDA also is required to consider the potential environmental effects of the actions to be proposed. The agency believes that issuance of any regulations as a result of this advance notice would not significantly affect the quality of the human environment. FDA requests comments on the potential environmental consequences of implementing the policy discussed in this notice. References The following information has been placed on display in the Dockets Management Branch, Food and Drug Administration, and may be seen from 9 a.m. to 4 p.m., Monday through Friday. 1. Tumutis. L. "The Value of Long-Term Testing for Implementation of Primary Prevention," in "Origin* of Human Cancer," Edited by Hiatt, H. H., et ah. Cold Spring Harbor Laboratory. NY. pp. 1339-1357,1977. 2. Occupational Safety and Health Administration, Final rule to add 29 CFR Part 1990. "Identification. Classification, and Regulation of Potential Occupational Carcinogens." Federal Register, January 22, 1900 (45 FR 5001). 3. Criesemer, R. A. and C. Cueto. Jr.. Toward a Classification Scheme for Degrees of Experimental Evidence for the Carcinogenicity of Chemical* for Animals," in "Molecular and Cellular Aspect* of Carcinogen Screening Tests," Edited by Montesano, R,, et al,, International Agency for Research on Cancer, Lyon, France, pp. 259-281,198ft 4. National Toxicology Program. Fiscal Year 1980 Annual Plan. Waahington, DC: Department of Health, Education, and Welfare. Public Health Service, 1980. 5. Caylor, D. W. and R. E. Shapiro. "Extrapolation and Risk Estimation for Carcinogenesis," Advance* in Modem Toxicology, Volume 1. Part Zi "New Concepts in Safety Evaluation," Edited by Mchtman. M. A., et ah, John Wiley h Sons. New York, NY. pp. 85-87.1979. 6a. Crump, K. S., "An Improved Procedure for Low-Dose Carcinogenic Risk Assessment From Animat Dsta." Journal of Environmental Pathology and Toxicology, in press. b. Environmental Protection Agency. Notice of Availability; Water Quality Criteria Documents, Federal Register. Appendix C. November 28.1980 (45 FR 79318. 79347). 7a. Hoel, D. G.. et ah. "Estimation of Risks of Irreversible. Delayed Toxicity," Journal of Toxicology and Environmental Health, 1:133151,1975. b. Food end Drug Administration, Proposal to amend 21 CFR Parti 70, 500.514 and 571. "Chemical Compounds in Food-Producing Animals," Federal Register, March 20,1979 (44 FR 17070). c. Caylor, D. W. and R. L. Kodeli, "Linear interpolation Algorithm tor Low Dose Risk Assessment of Toxic Substances." Journal of Environmental Pathology and Toxicology, 4:305-312,1980. 8u. Williams. C. M.:a. Leff. and J. H. Weisburger. "A Species to Species Comparison of Carcinogenicity Data, with Human Extrapolation," Final Report to the National Institute of Environmental Health Sciences (Contract No. l-ES-6-2130), July 1978. b. National Academy of Sciences. "Pesl Control: An Assessment of Present and Alternative Technologies." Contemporary Pest Control Practices and Piospects. Vol 1. pp 54-101. c. Crouch. F. and R. Wilson. "Interspecies Comparison of Carcinogenic Potency." /ouruiil of Toxicology and Environmental Health. 51095-1118.1979. Interested persons may, on or before July 1.1982. submit to the Dockets MEV-144063 v ~7 ATTACHMENT C MEV-144064 T Chapter I--Feed and Drug Administration 5 174.5 11*3.345 Sodium methyl sulfate. Sodium methyl sulfate may be pres ent In pectin in accordance with the followinc conditions. <a> It is present as the result of methylation of pectin by sulfuric acid and methyl alcohol and subsequent treatment with sodium bicarbonate. (b) It does not exceed 0.1 percent by weight of the pectin. 1173-395 THfluoronxthan* sulfonic acid. Trifluoromethane sulfonic acid haa the empirical formula CFiSOiH (CAS Res. No. 1493-13-6). The cata lyst (Trifluoromethane sulfonic acid) may safely be used in the production of cocoa butter substitute from palm oil (l-palmitoyl-2-oleoyl-3-stearln) (see f 184.1239 of this chapter) in accord ance with the following conditions: (a) The catalyst meets the following specifications: ApDcarmncc. Clear liquid. Color. Colorless to amber. Neutralisation equivalent. 147-151. Water. 1 percent maximum. Fluoride ion. 0.03 percent maximum. Heavy metals (as Pbi. 30 parts per million maximum. Arsenic (as Asi. 3 parts per million maxi mum. (b> It is used at levels not to exceed 0.2 percent of the reaction mixture to catalyze the directed esterification. <c> The esterification reaction is quenched with steam and water and the catalyst is removed with the aque ous phase. Final traces of catalyst are removed by washing batches of the product three times with an aqueous solution of 0.3 percent sodium bicar bonate. id) No residual catalyst may remain n the product at a detection limit of 0.2 part per million fluoride as deter mined by the method described in "Offlcal Methods of Analysis of the Association of Official Analytical Chemists." sec. 25.046. 12th Ed. (1973), which is incorporated by reference. Copies are available from the Division of Food and Color Additives. Bureau o Foods (HFF-330). Food and Drug Administration. 200 C St. SW.. Wash`ftgton. DC 20204. or available for in spection at the Office of the Federal JJegister. 1100 L St. NW.. Washington. DC 20408. [43 FR 34237. Nov. 11, 1974. u amended at 47 FR 11839. Mar. 19.19121 PART 174--IMDIRICT POOD ADOlTtVISi OINRRAL 1174.3 General provisions applicable to Indirect food additlvee. (a) Regulations prescribing condi tions under which food additive sub stances may be safely used predicate usage under conditions of good manu facturing practice. For the purpose of this part and Pam 173.176. and 177 of this chapter, good manufacturing practice shall be defined to include the following restrictions: (1) The quantity of any food addi tive substance that may be added to food as a result of use in articles that contact food shall not exceed, where no limits are specified, that which re sults from use of the substance In an amount not more than reasonably re quired to accomplish the intended physical or technical effect in the food-contact article: shall not exceed any prescribed limitations: and shall not be intended to accomplish any physical or technical effect in the food itself, except as such msy be permitted by regulations in Pam 170 through 189 of this chapter. (2) Any substance used ss s compo nent of articles thst contact food shall be of a purity suitable for its intended use. (b) The existence in the Subchapter B of a regulation prescribing safe con ditions for the use of a substance as an article or component of articles that contact food shall not be construed to relieve such use of the substance or ar ticle from compliance with any other provision of the Federal Food. Drug, and Cosmetic Act. For example, if s regulated food-packaging material were found on appropriate teat to impart odor or taste to a specific food product such as to render it unfit within the meaning of section 402(a)(3) of the act. the regulation would not be construed to relieve such use from compliance with section 402(a)(3). (c> The existence in this Subchapter B of a regulation prescribing safe con ditions for the use of a substance as an 121 VEV-144065 S U5.105 Tltlo 21 Food and Drugs article or component of articles that contact food shall not be construed as implying that such substance may be safely used as a direct additive in food. <d) Substances that under conditions of good manufacturing practice may be safely used as components of arti cles that contact food Include the fol lowing. subject to any prescribed limi tations: (l> Substances generally recognized as safe In or on food. (2) Substances generally recognized as safe for their intended use in food packaging. (3) Substances used in accordance with a prior sanction or approval. (4) Substances permitted for use by regulations in this part and Parts 175, 176. 177, 178 and 1 179.45 Of this chap ter. (Sec. 409. 72 Stst. 1789. 17SS. u amended (21 U.S.C. 34S. 37D) [42 FR 14S14. Mar. IS. 1977] PART 175--INDIRICT FOOD ADDU TIVISt ADHISIVI COATINOS AND COMPONINTS SmSmH A--(se*v*e) iMbeert I leSMeew far Uie Ody * Ccpemwti af Adhedvae S#C* 175.109 Adhesives. 175.125 Pressure-sensitive adhesives. guAmAeS S*a Usa mm pAmwwdfe^Wfta --A P-- WW 175.210 Acrylate ester copolymer coatings. 175.230 Hot-melt stnppable food coatings. 175.250 Paraffin (synthetic). 175.250 Partial phosphoric acid esters of polyntir rttln 175.270 Poiytvtnyl fluoride) resins. 175.300 Resinous and polymeric coatlnss. 175.320 Resinous and polymeric coatlnss for polyolefin films. 175.350 Vinyl acstate/crotontc acid copo lymer. 175.310 Vlnylldene chloride copolymer coatlnss for nylon film. 173.345 Vlnylldene chloride copolymer coatings for polycarbonate film. 175.350 Xylene-formaldehyde resins con densed with 4.4 -isopropyiidenedlphenoi- eplchlorohydrtn epoxy resins. 175.190 Zinc-silicon dioxide matrix coat lnss. Atmtoamr Sees. 400. 701. S3 Stat 10551055 as amended. 72 Stat. 17S5-17SS u amended (21 U.S.C. 345). unless otherwise noted. Sooact: 42 FR 14334. Mar. IS. 1977. unless otherwise noted. Subpart A--[Kosorvod] Subpart 1--Subatancao far Use Only as Ceaipenents af Adhesives 175.105 Adhesives. (a) Adhesives may be safely used is components of articles Intended for use In packaging, transporting, or holding food In accordance with the following prescribed conditions: (1) The adhesive is prepared from one or mor of the optional substances named In paragraph (c> of this section, subject to any prescribed limitations. (2) The adhesive is either separated from the food by a functional barrier or used subject to the following addi tional limitations: (I) In dry food*. The quantity of ad hesive that contacts packaged dry food shall not exceed the limits of good manufacturing practice. (II) In fatty and aqueous foods, (a) The quantity of adhesive that contacts packaged fatty and aqueous foods shall not exceed the trace amount at seams and at the edge exposure be tween packaging laminates that may occur within the limits of good manu facturing practice. <b> Under normal conditions of use the packaging seams or laminates will remain firmly bonded without visible separation. <b> To assure safe usage of adhe sives. the label of the finished adhe sive container shall bear the statement "food-packaging adhesive". (c> Subject to any limitation pre scribed in this section and in any other regulation promulgated under section 409 of the set which prescribes safe conditions of use for substances that may be employed aa constituent! of adhesives, the optional substances used in the formulation of adhesives may Include the following: (1) Substances generally recognised as safe for use in food or food packag ing. 122 > ATTACHMENT D 44067 "GOOD MANUFACTURING PRACTICES" CRITERIA FOR PLASTIC RESINS "PRIOR SANCTIONED" UNDER THE FOOD ADDITIVES AMENDMENT OF 1958 We smtest Hut m l>is tilletii mtn im urmut telenice aterial. ' inarch. 1963) VEV-144068 PREFACE This Bulletin has been prepared by the Food Packaging Materials Committee of The Society of the Plastics Industry, Inc. in the hope that it will provide helpful informa tion in a compact form to those who supply plastics materials for food packaging and processing applications. A great deal of the credit for this work is due the Technical Information Subcommittee of the Food Packaging Materials Committee for the more than two years of effort that went into preparation of this publication. The data included was screened and placed in its present form by the Subcommittee after all known pro ducers of the plastics covered were painstakingly contacted and afforded a full opportunity to supply information about their products. Thereafter this material was circulated in draft form and ultimately finalized as it now appears. We also acknowledge with gratitude the cooperation given the Society by the Department of Agriculture in supplying the two letters reproduced herein attesting to the acceptability of the listed materials for use in Federally inspected meat and poultry plants. It is believed that this Bulletin will prove a valuable reference for all those making food packaging or processing materials or components thereof. Its publication is not intended however, as any express or implied warranty as to the general fitness of the products covered for all specific uses; this is a matter which must necessarily be left to case-by-case determination dependent on the nature of each intended application. The Society of the Plastics Industry, Inc. y "GOOD MANUFACTURING PRACTICES" CRITERIA FOR PLASTIC RESINS "PRIOR SANCTIONED" UNDER THE FOOD ADDITIVES AMENDMENT OF 1958 On July 22, 1960, the Food and Drug Administration, by means of a letter directed to our Counsel, reaffirmed the "prior sanctioned" status of the major plastic polymers now in use in food contact applications, and not specifically covered by Food Additives Regulations. This landmark letter had this effect as a result of its reference to an article entitled "Food Packaging" by Dr. A. J. Lehman, Chief of FDA's Division of Pharmacology, which appeared in a widely disseminated publication in 1956 *. In the introduction to this article Dr. Lehman stated: "That class of materials loosely referred to as plastics are currently being exploited widely for food-packaging purposes..... Our purpose now is to discuss these new developments and bring up-to-date our listing of materials for food packaging which we have reason to regard as acceptable." The reaffirmation of the status of these materials is of vital significance to the plastics industry because materials recognized as "prior sanctioned" are exempt from the coverage of the Food Additives Amendment of 1958 and, hence, clearance of these substances by means of Food Additives Regulations is not required. For the information of those who may not have seen copies of the FDA letter, a copy is included in this publication. Reference to the letter, as well as reference to the Lehman article upon which it comments, will demonstrate that the approvals or "prior sanctions" are limited by the condition that the "resins" covered be made "in accordance with good manufacturing practice, for use in food" applications. Since criteria for determining whether a polymer on the Lehman List has been made in accordance with "good manufacturing practice" have not been available in a single source heretofore, the SP1 Food Packaging Materials Committee has collected data from which it has been able to compile definitions based on materials characteristics so that the industry in general may have a more concrete concept of what is involved insofar as individual polymers are concerned. The listings and descriptions in this publication are designed for this purpose. As will be noted, the listings herein deal only with those basic resins or polymers covered in the Lehman List.*1 There are undoubtedly other materials which are "prior sanctioned" by virtue of letters sent to individual companies prior to the enactment of the Food Additives Amendment. This publication does not purport to cover other such materials, nor is it intended to encompass the adjuvants or optional ingre dients which may be employed in finished or intermediate formulations of products made from the Lehman List, or any other substances. 1 Lehman, A. J., Quart Bui:. Assn. Food and Oni| Officiala of the United States, XX, No. 4, p, ISO-168, October, 1956. 1 It should be noted that the following resins originally listed by Dr. Lehman era not included in the coverage of this publication: Condensate dimethyl terephthelate mid ethylene glycol (Regulation 121.2324) Polyvinyl alcohol (Petition FAP No. 209) Polyethylene (high and low preaaure typea)(Regulationa 121.2308 and 2510) Regenerated celluloae (Regulation 12L 2307) Torephthalic acid-ethylene glycol copolymer (Regulation 12L 2524) Styrene butadiene copolymer (Petition FAP No, 602 end 662) Styrene iaobutylene Polystyrene (normal mid rubber modified)(Petitiona FAP Nos. 602 and 662) All of these polymers with the exception of styrene isobutylene are the subject of new Food Additives regulations or pending petitions for regulations to that it appeared inappropriate mid unnecessary to further define "good manufacturing practices" for them. In those esses where Petitions are in the pending status, the materials involved remain "prior saictioned" of course. The sane may be said to be true as to thosonow coveted by Regulations sines ths Food ad Drug Administration has not exercised its power to withdraw any such sanctions. 1 VEV-144070 Generally speaking, such adjuvants or optional ingredients may be employed provided they are among those considered "prior sanctioned", "generally recognized as safe", or for which an appropriate Food Additives Regulation exists. Likewise, if it can be concluded that a substance may not "reasonably be expected to become a component of food," i.a. it will not transfer from a plastic to food, it may be used in conjunction with any of the polymers. The following paragraphs describe the end characteristics of significance to a determination of whether the resins or polymers covered are made in accordance with good manufacturing practice for food packaging or processing applications and are, therefore, "prior sanctioned.^* 1, Polyvinyl Butyrol Viscosity is 2-500 cps, determined as follows: Classification by Viscosity Level _____ and Hydroxyl Content Solvent Concentration IliV-HiH HV-LoH LV-H11 LV-LH 50-500 cps 20-200 50-500 100-500 up to 50 cps up to 20 up to 50 -- -- up to 100 Methanol 7.5% Methanol 6.0% Ethanol 5.0% Up to 30% unreacted polymeric hydroxyl (calculated as polyvinyl alcohol) is generally present and permits crosslinking via phenol- or urea-formaldehyde resins or by other means. Some grades contain up to 30% of polyvinyl acetate, as determined by ASTM Method D-139<*58. 2. Polyvinyl Chloride Maximal volatility is not over 3.0% (1 hr. at 105C) and inherent viscosity not less than 0.35 as determined by ASTM D-1243-60, Method A. 3. Polyvinyl Acetate Maximal volatility is not over 3.0% (45 min. at 135C) and viscosity not less than 1.35 centipoise at 20C in benzene solution (4.3 gm. resin/50 ml.). 4. Polyvinyl Chloride-Acetate Maximal volatility is not over 3.0% (1 hr. at 105C) and inherent viscosity not less than 0.30 as determined by ASTM D-1243-60, Method A. 5. Polyvinylidene Chloride Viscosity is within the range of 0.50-1.50 centipoise at 120C (2% solution in o-dichlorobenzene [which is Cow BS-SB-B method in Appendix A. Questions on the method, especially those on item 5 "Visco meter Calibration" may be referred directly to Dow] ). .6 Cellulose Acetate Maximal free acid of cellulose ester "flake" after washing and drying is no more than 0.02% by weight (ASTM D-817-57)*. 7. Butadiene-Acrylonitrile Synthetic Rubber In plastics, butadiene-acrylonitrile (as is butadiene styrene) is used to modify or "toughen" other polymers, principally polystyrene. The rubber's contribution to the resultant plastic depends on th molecular weight and monomer ratio of the rubber, measurable by ASTM D-1646-59T and Kjeldahl determination for nitrogen content (indicative of bound acrylonitrile in the rubber) respectively. There need be no restrictions on these factors because the commercially available butadiene-acrylonitrile rubbers have not significant extrectables. 1 The ester "flak*" is never used other than as a component of a formulation, all of the components of which must someway be cleared or exempt from the Food Additives Amendment as suggested in the opening paragraphs of this article. The above limitations are believed to apply to cellulose acetobutyrate and cellulose proprionate; as to cellulose nitrate, a nitrogen content of 10.9% to 12.2% was prior sanctioned for food packaging purposes. 2 VEV-144071 A .8 Ethyl Cellulo Viscosity is within the range of 3-5000 cps (25C, 5% solution in 80:20 toluene: ethanol [ASTM D-91450]). Kthoxyl content is within the range 43-51.5%. 9. Rubber Hydrochloride Chlorine content is within the range of 30-32%. One mil films of the unmodified product are to have a volatile content of no more than 0.5% (4 hrs. at 75C) and a hexane extract of no more than 2.0% (2 hrs. reflux). 10. Polyester Resin-Ethylene Terephtholote and Ethylene Isophthalate Maximum volatility is no more than 0.5% when heated at 60C for 12 hrs. at a pressure less than 10 mm of mercury. Hexane-soluble content (2 hrs. reflux) is less than 1.0%. .11 Butadiene-Acrylonitrile-Styrene Total residual monomer is primarily styrene and is less than 1.0%, determined by the SPI method (Appendix B)* for rubber modified polystyrenes, using chloroform as solvent. 12. Polymer of 2-Chloro-Butodiene Maximal volatility of the dried coating is not over 0.5% (4 hrs. at 75C) and hexane extraction is not over 2.0% (2 hrs. reflux). 13. Polymer of Melemin^Formaldehyde The resin is cured at selected time-temperature conditions which will evaporate solvents and produce insolubility by crosslinking. 14. Polymer of UreoFormoldehyde (Includes unmodified resins, and those reacted with selected amines, to develop solubility in water sufficient for application purposes.) The resin is cured at selected time-temperature conditions which will evaporate solvents anti produce insolubility by crosslinking. 15. Polymer (Fluid) of Dimethylpolysiloxone The polymer is clear, has a maximal acid number not exceeding 0.5 (MIL-S-21568A, paragraph 4.6.8) and has the following additional physical properties: Nominal Viscosity (Centistokes) __________ Maximum Volatility (%) (48 hrs. at 200C) (ASTM D-972-48T) Maximal Color (ASTM D-1209-54) 3504 2 40 4Prior sanction* far higher viscosity types are believed to exist. *The shaking-time requirement in Section 6b (page iv) should be extended from 3 to 24 hours to insure complete removal at styrene monomer. 3 44072