• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Human Food Safety Issues

Dr. Kornelia Grein

DR. GREIN:  Good morning, ladies and gentlemen, Mr. Chairman.  Firstly I wish to congratulate the organizers for this workshop for this event.  I think it is an excellent forum for exchanging the approaches, the ideas and the problems, how to overcome the difficulties in having drugs available to treat minor species and minor uses in major species, which is definitely a problem throughout the world as we have seen yesterday.  Then I want to thank all of the organizers for having invited me to speak here on the issue of human food safety. 
The points I want to address, what are the safety issues, and we have to go back.  I want to report what are the approaches that is being taken in the EU on this, and then I want to address specific issues then --- requirements for minor species, minor uses, and extrapolation of maximum residue limits.
The issue under consideration is how the consumer of food originating from animals which has been treated with drugs can be protected from harmful residues in food if the animals concerned are minor species or if the treatment concerned a minor use of major species.  This protection is achieved the same as this major species and major uses, through the establishment of maximum residue of tolerances and application of withdrawal periods.  The question is then what are the data requirements and the scientific approach on how to assess the data.  Should it be identical as for the major species and major uses, or whether there can be any adaptation to this approach for the MUMS products?  Certainly I think this is an issue where nobody would question this: the overlying principle for any such considerations must be that the consumer health must not be put in jeopardy.
On the next slides I want to present to you the situation in the EU, the approach that we take in the EU, and on the first slide I come back to the definition of minor use and minor species, as you have seen yesterday already also in the presentation from Peter Jones.  One point is clear:  In the EU we do not have a definition of minor species and minor uses in the legislation.  We have only a definition in CVMP guidelines.  The CVMP, I believe everybody is familiar with the term, it is the Committee for Veterinary Medicine Products, and since we have a new legislation Committee for Medicinal Products for Veterinary Use.  It is the scientific committee which is part of the EMEA, and it meets once per month for three days and it is responsible for the scientific evaluation of the centralized marketing authorizations, the maximum residue limits in the EU and scientific advice.  It also establishes the guidelines that do support the legislation, and the CVMP has the support of working parties for this task.  The experts that are members of the CVMP come from all the EU member states. 
So, the definition given by the CVMP in the guidelines establishes what is a major species, and it is cattle, sheep, pigs, chicken, and salmonidae.  All the other remaining species become minor then.  The CVMP, when this list of minor species was established, and it was a couple of years ago, looked to consumption data in the EU, consumption figures for food.  Cattle and pig meat presents 94 percent of the meat consumption in the EU.  Sheep meat is actually much, much smaller, two to four percent.  However, regionally sheep meat is a major food commodity --- and we cannot use the average across the whole of Europe.  We have to ensure that there are regions that need to be protected in the same way.  That is why sheep meat in the EU is a major species.  Fish consumption is 25 percent compared to meat.  Minor species then are other ruminants, particularly goats and deer.  When we look to other avian species those that are minor are duck, goose, game birds and so on.
I just want to show you the guidelines.  I don’t want to go into detail now, but merely list them for reference.  There are two guidelines actually which address the point.  The first one is this Note for guidance on the establishment of maximum residue limits for minor animal species.  From the number you see, this goes back to the year 1997, so it is a couple of years ago already that the CVMP looked at that aspect in respect to maximum residue limits.  Actually, this wasn’t even the first guideline.  Two years before there was already a preceding guideline on that matter, as it was always considered important to have MRLs,  even at the time, for minor species, in order to have adequate licensed products on available and also considering the off-label use because to the control of residues it is important that MRLs are available.
The other guideline, we looked to the same issue a couple of years later.  As you can see again from the number, this guideline was proposed in the year 2000.  It is called Note for guidance on risk analysis approach for residues of veterinary medicinal products in food or animal origin.   It reviewed again the assessment approach for setting MRLs and extrapolating them to other species and the details of that guideline are presented later.
The approach in the EU to establish MRLs is scientifically very much similar as to the JECFA or for setting tolerances as outlined by Dr. Baynes just before in the previous speech.  We have a battery of toxicity tests.  The NOEL has been established.  The lowest NOEL is being used by applying an uncertainty or safety factor to establish an ADI.  We have a residue data package, pharmacokinetics data, residue depletion data, and we use a standard food basket and on this basis the MRL is being derived, but [the NOEL should not -- or] the daily intake should not exceed then the ADI.  Regarding the target tissues we are looking at muscle, fat, or fat on the skin for pigs and poultry, liver, and kidney, and where appropriate also eggs and honey. 
The first MRLs that were established in the EU go back to the early ‘90s.  We were following the JECFA approach established for all food producing species.  These were penicillins, sulfonamides and tetracyclines.  Later on following again JECFA, there was a distinction between the species being made.  So then also species-specific data were requested on the residues and analytical methods for the species we are looking at.  So we have then later on separate MRLs for cattle, sheep, goats, poultry.  It would be a separation between chicken, turkey, and so on. 
Later on, when all the MRL assessments had been completed in the EU for all substances -– I have to say a little bit about procedure later on this whole approach was being reviewed again in this guideline I mentioned already before. 
The situation in the EU with regard to the MRL procedure, and this may be different to other countries, is that we have a separate procedure to establish MRLs.  MRLs are a precondition for a marketing authorization for any veterinary medicinal product for food-producing animals.  There is a separate legislation, and since this legislation exists, it dates back to the year 1990 and came into effect in ‘92, it has always been a kind of centralized procedure.  There is only one MRL in the EU for a substance though, we have the different member states in the EU.  It was never in the EU legislation that the individual member stats could establish separate MRLs.  It makes sense.  We have to ensure the same level of safety for all consumers throughout the European community and the trade issue.  We cannot have free trade of food originating from animals within the EU with different MRLs, this simply does not work.  So we had always this precondition, even if then in the next step marketing authorizations for the drugs would be issued by individual member states.  They would establish the withdrawal periods fro the products ensuring compliance with the MRLs.
The legislation then lays down the data requirements in an annex to the legislation and this is a  list of the different endpoints, and it is a list you know from the normal data package for safety.  That is why we require, to set MRLs, pharmacological studies, the set of toxicity studies, then a residue file comprising depletion studies and an analytical method.  Nowadays this toxicity data package would be according to the VICH requirements which have recently been agreed, but everything has yet to be fully implemented because the two last guidelines were only adopted this year, and for one other guideline the developmental toxicity guideline we need actually a change in the legislation, but this is underway.  That is the situation for major species.
In the guideline on minor species from the year ‘97, the CVMP looked at the time if one could have a certain reduction of data requirements if MRLS were requested for minor species only.  The approach that had been agreed at the time was that for example no pharmacokinetic study would be required for this purpose then, but as a kind of surrogate studies to establish the oral bioavailability in place of pharmacokinetics would be accepted.  In respect to repeat dose toxicity, at the time, the approach was: two species 28 days tests or one species 90 days test would be accepted in place of two 90-day tests for major species.  For reproductive toxicity, teratogenicity, also certain reduction with them:  One generation reprotox study instead of two generations.  For immunotoxicity, which actually hardly ever was relevant for the MRLs dossier that would not be required, but tests on other relevant effects.  Mutagenicity, there was no deviation at all and also for the residue file, that should be the same as for major species.
As Dr. Jones mentioned yesterday, CVMP is at the moment looking to review entirely --- for any of the data requirements for -- in the --- dossier, also --- any possibility to adapt the data requirements for minor use and minor species.  In this context also the CVMP, the safety working party looks again to the requirements for the MRLs.  What I will show you here is just, are really considerations, ideas, thoughts of the working party.
  (Adjusting the microphone.)
Okay.  So these are no final conclusions.  These are just considerations at the moment, but at this event here I thought it might be useful to share the results with you.  The outcome can in a couple of months be different, but it might be useful to see what has been discussed.  So, with respect to repeat dose toxicity, experts have said a 28-day study doesn’t make sense.  One needs a longer term study, and there was the question: Is one study, a 90-day study, sufficient?  Would one need not two different species?  Regarding reproductive toxicity, teratogenicity, they also said a one-generation reprotox study is not of much of value.  If one carries out a study, it should be a two-generation study.  Then it was discussed,  can one really justify having no study at all available.  For teratogenicity, embryotoxicity, the VICH approach, which is a tiered approach, was considered as a good approach by the experts.  Anyway, the new VICH data set would be required.  In respect to carcinogenicity, it would support not to require to test, if a criteria set would apply like, for example, if there is no analogy in chemical structure with a well-known carcinogen negative mutagenicity tests and no carcinogenicity signs have been seen during toxicity testing.  In respect to a routine analytic method, it was considered necessary to have it largely similarly validated as for major species because it is required for the residue surveillance.
Regarding marketing authorizations, for the toxicity data would be the same consideration as for the MRLs.  In a way, it is the same sponsor, they have anyway the data set available and, so the ADI has been established.  There is no need for having another consideration.  We would also look to: Are there any possibilities now in respect to withdrawal periods, and this would be now more for cases where we have already a maximum authorization for example for a major species –- could one extrapolate withdrawal periods?  Also the question was asked what is it with analytical methods.  Is always a fully-validated method necessary if one has already a product available for a major species?  So these are questions that are now under discussion; and we expect that these guidelines may be finalized by the beginning of next year and they will be then issued for consultation in the next year, and we will be interested to see the comments on this.
Now I want to come to the issue of extrapolation of MRLs.  I said before, the CVMP reviewed the whole approach, how specific MRLs have to be at the time.  The assessment for all old MRLs has been carried out.  To explain what has happened is that when this legislation on MRLs came into place in ‘92, it was said at the time that for any new product that comes on the market MRLs must be established before.  However, of course many products were on the market, and for those it was a transition period originally until the year ‘97, which was extended then until the year ‘99, where all -- for all pharmacologically-active substances contained in veterinary medicine products for food-producing animals must have been assessed and MRL been established.  Otherwise, the products would have to disappear on the market.  So the EMEA assessed at the time over 700 substances for which we had to receive dossiers first.  For around 100 substances, we could at the end not establish an MRL because the data were insufficient.  Sometimes, for a majority, the applications were withdrawn after we had sent to the companies a so-called list of questions, or incomplete letters as it is also being called, or sometimes the responses were still insufficient, and we had not sufficient data.  For more than 100 substances which MRL values were established, around 500 substances where considered that no MRL is necessary, and we put them in a list which sometimes contains clarifications for which species, the supply of which route of administration this would apply.  This is one of the lists, the annexes.  There are certain substances, number of 11, that are completely forbidden in the EU, but also any substance that is not on this kind of positive lists are equally forbidden to be used in veterinary drugs for human food consumption.
So we had a wealth of data at that time then available from all the substances, and we looked again how to the approach in the future in respect to extrapolations.  So we arrived then at this list that MRLs could be extrapolated from major ruminants to all ruminants if possible, from major ruminant milk, that means cattle meat, to sheep milk, goat milk, major monogastric mammals, that means from pigs for example, then to horses, rabbits, chicken, to poultry and from salmonidae to all other, fin fish.  Also in addition there could be if identical or similar MRLs were derived in cattle or sheep, pigs and chicken, MRLs can be extrapolated to all food-producing species.
We would need for the extrapolation to minor species a confirmation of the marker reside.  So a certain amount of residue data, but certainly no radio labeled study, and approval of the applicability of the analytical methods.
The guideline was implemented then, and we had at the time very much the call of the veterinarians, the industry, to make automatic extrapolation, looking -- opening all the dossiers again, looking for data available.  We did this only for those substances where we could do an extrapolation to all food-producing species.  We checked in particular the analytical methods; are they are applicable?  This list of substances came out of this.  There were three substances more, but they had really insufficient analytical methods to allow for this approach. 
For the remaining possible extrapolations, the CVMP would of course apply this approach for any new application that would be in the process, but we wanted -- had the plan that we would wait for applications from companies with dossiers, paying the fee, and would undertake then only the work if there is really a sponsor out there who wants to bring the product on the market.  Anyway, not unexpectedly, because we know also there is not much interest on the reasons we heard yesterday from companies to develop additional claims for minor species, we did not have many applications.  Certainly we still wanted to have products on the market, so the CVMP considered that, you know, of course MRL is only an MRL, not a marketing authorization, but it could extend and incentive for industry.  So we went ahead and tried to look at other possibilities, looking where there are really essential substances.  Areas, species, where we know there is a lack of products, and we would make then a list and try to extrapolate MRLs to these species where they are needed without an application, without any fee. 
The first kind of trial where we applied this approach was in respect to sheep and goats, particular milk-producing animals where the veterinarians in Southern Europe particularly said there are no products available, and in particular for ectoparasiticides.  There was obviously a real lack of any product that is authorized, so we established a list of substances for which MRLs could in principle be extrapolated.  We wanted to have only the old non patens products, not favor any company, a single company who could benefit on this, and of course we looked if the analytical methods would be available.  We submitted then a list for consultation, and this is then after consultation the list of substances for which these recommendations for extrapolations could be made.  I believe they have been transferred into legislation in the meantime.
For the future we want certainly to look for any Annex II substances at some time.  There could be the one or the other extension possible.  If needed, we would look to this without fee if a company wants to apply for an extension of Annex entries substances without MRLs I have to say.

Certainly any application from a company or a sponsor who wants to make an extension would be considered, but also we may look to other extensions for essential substances.  Maybe chicken or rabbits, I think that is possible if we have the data in-house that could support this.
To sum up, reduction of data requirements it seems for MRLs and marketing authorizations that are exclusively used for minor species, are only possible to a limited extent.  We want to ensure consumer safety.  If MRLs or marketing authorizations are there for major species, there seems some extrapolation possible, certainly with the MRLs, but also possibly in some cases with regard to withdrawal periods and analytical methods.  Thank you.
DR. WEBB:  Well, we are just about to go on a break.  We would ask you to be back at five minutes past 10:00. 
Just one thing, I have been asked by several people about the absence of a representative from China.  This was all arranged, and the individual got promoted and was immediately given other jobs and had to drop coming to this meeting.  The alternate person, it was too late for them to get a visa from China to come.  So that is the reason.  However, to show that -- they didn’t do this to show this, but just a note to you, they are interested in continuing this dialog.  They actually have joined FARAD, so there is going to be continuing international contact with them. Public comments, please, if you have any, register at the front desk.  Thank you, and back at five past 10:00.
  (Whereupon, a break was taken.)