CVM’s announcement that the memorandum of understanding (MOU) between CVM and AAFCO would not be renewed should not be a surprise to anyone, because eventual dissolution of the arrangement was obvious from day one with the first MOU issued in 2007.
How did we get entangled in such an imbroglio? In 1906, federal (i.e., interstate) regulation addressing control over human food was just being initiated with the Food and Drug Act, but federal regulation addressing animal food or animal food ingredients would not come for another 50+ years. So, why would eventual federal intervention into animal feed become important? As you probably know, in the early part of the 20th century, the individual states wielded power more frequently than the federal government and many of the states had already passed laws of their own regulating feed and ingredients marketed within the state. Because this patch work of state laws resulted in inconsistencies between states, individual state regulators formed a private company, the Association of American Feed Control Officials (AAFCO), whose mission was to resolve these inconsistencies and level the playing field between the states. The task of the AAFCO was to produce a feed ingredient definition for addition to the AAFCO Official Publication (OP) and in a format that would be acceptable to all states as a model legislative bill which could be submitted to the many state legislatures. If accepted by the legislatures, the bill would be passed on to the governor for signing and become the law of the land (i.e., the state).
This system of state control over animal feeds worked well until 1958, when the Food, Drug and Cosmetic Act was amended to include a mandate that all human and animal food and their ingredients should be safe for their intended use prior to marketing (i.e., pre-market approval). The amendment decreed that all food and feed ingredients that are not generally recognized as safe (GRAS) are food additives, the latter of which must be approved by the FDA via a food additive petition. On the other hand, an ingredient could be found GRAS by two means: either by experts (deemed such by their training and experience) or if demonstrated as already having been in use in food prior to January 1, 1958 (i.e., a presumption of safety as the result of experience).
With the possible exception of feed ingredients in common use prior to January 1, 1958, the requirements for GRAS eliminated any legitimate legal standing for ingredients in the Official Publication because AAFCO failed to meet at least three essential criteria for compliance: (1) the ingredients had never been declared GRAS, but only listed in the Official Publication (OP); (2) nor was safety in use ever declared in the OP and; (3) the persons approving entry into the OP never verified their bona fides as “experts by training and experience”. That is, the substances in the OP were perfectly legitimate within the states where approved by state law, but because the ingredients were not compliant with federal law, the ingredients could not enter into interstate commerce, becoming violative for use as a feed ingredient as soon as they left the state of origin.
So, what is the difference between an ingredient approval by AAFCO or organizations like the Flavor and Extract Manufacturers’ Association (FEMA) or the Enzyme Technical Association (ETA)? While AAFCO failed to meet three threshold criteria for compliance, the same is not true for FEMA or ETA. Each time FEMA produces a list of GRAS ingredients, the ingredients are designated as being GRAS and the list is published in the widely circulated Food Technology magazine, the safety in use of the ingredients is described in a scientific journal or submitted to FDA, and the designating experts – the FEMA Expert Panel – has very credible bona fides. For ETA, this organization submits its GRAS conclusions to FDA for the GRAS Notification process – the GRAS Notification describes the safety in use of the enzyme, the Notification includes the identification of the experts whose procedures for determining safety were utilized.
How did FDA respond to the 1958 Amendment? While the human side for food ingredients (the Center for Food Safety and Applied Nutrition; CFSAN) was making rapid progress on compiling lists of GRAS ingredients, CVM saw the OP as its work already having been done – “don’t fix it if it ain’t broken”! This passive acceptance of OP ingredients was probably not the best of ideas, because not taking action against the ingredients in the OP as non-compliant, could be considered a de facto delegation of legislative power to a private entity by allowing AAFCO to set “metrics and standards”, a prohibition by the Circuit Court and confirmed by the Supreme Court in Department of Transportation v. Association of American Railroads.[1]
At the very least, CVM abandoned due process.
Over the coming years, CVM approved some feed additive petitions and following the GRAS review and affirmation process of human food ingredients by CFSAN in the late 1970’s, CVM adopted some of the substances vetted by CFSAN, although the vetting had been based largely on rodent data applied to humans, not target species. CVM also subsumed color additives approved by CFSAN as appropriate for animal use, although no list of approved colors or their uses (i.e., for food, drugs or cosmetics) or approved packaging ingredients have been published to the author’s knowledge. Ingredients included in the Code of Federal Regulations (CFR) by CVM, including food additives (21CFR572), GRASes (21CFR582) and GRAS affirmations (21CFR184) were subsumed into the OP.
As the system for animal ingredient review evolved, manufacturers could seek approval to market by going to the regulator in each state, obtain a feed ingredient definition from AAFCO and hope for approval by all states or, submit a food additive petition to CVM (which would be automatically included in the OP). At this point in time, CVM did not recognize GRAS as a legitimate pathway, despite a clear mandate in the Food Drug and Cosmetic Act (§201(s)) and any GRAS documents submitted to CVM went unanswered.
The year 2007 marked a seismic change for animal food regulation. The presence of melamine and cyanuric acid in pet treats produced a sickening of many pets and the death others from kidney failure. Anecdotally, there were more contacts with FDA over this poisoning than there had been for the last three human food contaminations – consumers were outraged! As a result, Congress passed FDA Amendments Act in that same year, which called for in part §1002(a)(1) [21 USC 2102] “ingredient standards and definitions with respect to pet food”. The use of the word “standards” was key to letting stakeholders know that food regulations would no longer be in the rearview mirror, but something that must be confronted head on. In response, AAFCO and others touted a “legislative fix” hoping Congress or the Commissioner would subsume all non-federally compliant ingredients in the OP (about 500 at that point in time) into an approved state – something that could not happen without an act of Congress – literally.
Despite protests from the old guard at CVM, Dr. Sundlof and his successors demanded quick action to remediate 70 years of neglect and mismanagement at the Center. A new era of transparency for the animal feed ingredient process was mandated. Actions taken included organization of a GRAS notification program based in part on the notification program at CFSAN and a publication describing the evolution of GRAS for the human side.[2]
Unfortunately, the GRAS notification program got off to a slow start, with few “no questions” letters issued. The slow start was a lack of understanding on both sides of what was needed, a prolonged review time (270 days) and a lack of transparency in the GRAS review process – namely, no opportunities for clarification or remediation of a submission as in the CFSAN GRAS notification program, resulting in a re-initiation of the entire process with another 270-day wait for a response.
How has AAFCO and the states responded to GRAS? AAFCO has given short shrift to GRAS determinations, even those having received a “no questions” letter from CVM. For those non-notified GRAS determinations, the individual states are reluctant to pursue these as an application for marketing in the state often pleading that the state does not have the resources to evaluate the GRAS. However, not to allow entry of the product in the state is a clear violation of the Supremacy Clause (Article VI, Clause 2) of the United States Constitution, which establishes that federal laws take precedence over state laws. That is, for a substance to be GRAS, it is in conformity with federal law and theoretically, cannot be barred from sale in the state; although there may be some exceptions.[3]
Also, in 2007, the first Memorandum of Understanding (MOU) was generated. Many in industry were ecstatic to see a description of the process of animal feed ingredient review and an acknowledgement by CVM that it has and will continue to participate in forthcoming ingredient reviews. Many saw this as nearly a vindication of the past practices and interaction of AAFCO, industry and CVM and; that the process of review would continue in the sunlight. What people failed to recognize in the MOU was the expiration date of five years in the future. This expiration date was a not-so-subtle message that AAFCO and industry must get their act together and make the old entries (possibly as many as 500) in the OP and any new entries, federally compliant. Despite warnings of a few voices that the ingredients must become federally compliant and through an industry-sponsored GRAS review, the honeymoon would soon be over and all non-compliant substances in the OP could become subject to immediate recall as unapproved food additives (i.e., potential adulterants) and recall of any feeds containing an unapproved additive which had thereby become adulterated.
CVM has now announced that time has run out. The MOU will not be renewed. CVM has cited guidances indicating it has not totally abandoned AAFCO and the states, but the message is clear that a solution to lack of compliance has not been forthcoming even after 17 years of warnings that a day of reckoning may be coming. CVM is girding its loins in anticipation of a pending calamity the result of nearly 50 years of neglect and mismanagement. FDA does not need another embarrassment following a baby formula shortage, food contaminations and drug shortages; and expiration of the MOU is a clear indication that FDA wants to avoid another catastrophe.
Could the mandate to AAFCO and the manufacturers be clearer? On the CVM website, it clearly states: “To be legally marketed, an ingredient used in animal food (including pet food) must be the subject of an FDA-approved animal food additive petition (FAP) or be generally recognized as safe (GRAS) for the intended use in animal food.” [4]
CVM has offered two draft guidance(s) for Industry (GFI). The first, #293, indicates CVM will take no action on non-compliant ingredients listed in the OP for which FDA has no safety questions and for which the use of the ingredient does not cause the label to be false or misleading. It is doubtful no safety questions will arise, simply because there was no safety assessment in the first place (let alone for all the target species) and a simple search of the scientific literature could likely turn up enough published research to cast doubt on any ingredient – a nice way for a competitor or gadfly to spur CVM to action and maybe even a recall of feed containing the ingredient. Remember, the safety of an ingredient must be demonstrated and few, if any, of the non-compliant ingredients in the OP have any data supporting their safe use – those with no information supporting safety in use are still violative.
The second GFI, #294 the “Animal Food Ingredient Consultation (AFIC)” describes a procedure whereby a party would make its case as to why a non-compliant ingredient should not be declared violative. CVM proposes that if an ingredient is safe (but how is this proven without data?) and used in the manner approved by CVM, CVM would not intend to take enforcement action. This guidance presents a high degree of subjectivity allowed to CVM in determining its intention as to whether it would take action – obviously a roll of the dice in the absence of a strong case for safety, which is required by the law. That is, if CVM has denied a renewal of the MOU and publicly recognized that the non-compliant ingredients in the MOU are violative, why would CVM provide a free pass to the same ingredient as the result of a consultation? AFIC does not appear to be a “get out of jail free” card.
What to do now? There are few options on the table. The GFI do not appear to offer much relief, and a “white knight” legislative fix is not going to happen, mainly because there has been no agreement on safety by appropriately credentialed bodies. In the early 1970’s, when it was found that not all the original GRAS substances nominated as being safe in the early 1960’s, were, in fact, safe, CFSAN had the original list reviewed by a special panel, the “Select Committee on GRAS Substances” (SCOGS) in the early- to mid-1970’s. The results of the SCOGS reviews of these originally GRASed substances were turned over to CFSAN, which in turn, incorporated most of into the CFR as §184. For those which had a history of prior use, but very little or no data were relegated to §182 and whose continued use was subject to successful testing outcomes, although the data supporting their safety was little.
AAFCO, the states and industry could follow the example of CFSAN and several trade associations who have found themselves in a similar conundrum and engage in a vetting of the non-compliant ingredients. Burdock Group has experience in triaging and prioritizing long lists of ingredients for assessment. Once the initial order of work is determined, we would then launch into the critical phase of obtaining compliance for as many ingredients as economically feasible – it is unlikely that all or even a majority of the non-compliant ingredients would require target species testing, but it would have to gather enough data to show safety in use. All activities would be conducted on a confidential basis and costs for each ingredient or logical chemically similar groupings of ingredients would be shared by manufacturers according to average volume sold over the past five years – again on a confidential basis.
[2] G.A. Burdock and I.G. Carabin (2004) Generally recognized as safe (GRAS): History and description. Toxicology Letters. 150(1):3-18.
Burdock Group Consultants stands as a premier authority in food safety and regulatory compliance, delivering customized solutions across various industries. Boasting over 35 years of experience as a frontrunner in regulatory and scientific consulting, our physical presence in Orlando enables us to offer clients direct accountability and access to cutting-edge scientific databases. These resources contain the latest and most thorough information necessary to determine the steps required to introduce your product into the market.
