Скачать 0.86 Mb.
Dr. Richard Carnevale|
DR. CARNEVALE: Thank you, Dr. Fox. Good afternoon. I am Rich Carnevale, vice president of scientific and regulatory and international affairs for the Animal Health Institute.
Dr. Fox has provided you with an overview of the animal drug industry's concerns regarding the issue of antimicrobial resistance. At this time, I would like to comment on some of the more specific aspects of the framework.
In the introduction to the document, the CVM claims that new reports, particularly from Europe, have renewed concerns for the contribution of animal antimicrobial use to the development of resistance in food borne bacteria.
Several literature references have been cited to support their conclusions, and some of those have been commented on today. Their conclusions are that immediate action is necessary to change the regulatory approach and the approval of antimicrobials in food producing animals.
AHI believes that the citations provided do not in all cases represent new information, and moreover, do not provide the compelling scientific justification for such a significant change in animal drug approval requirements.
We would like to briefly comment on some of these publications as it builds our foundation for further comments on the specific framework proposals.
One of the key reports that is referenced in the document is that of Threllfall et al., from the Central Public Health Laboratory in Great Britain published in 1996. In a series of articles, the authors suggest that temporal increases in "resistance" levels of Salmonella typhimurium, Determinant Type 104, are directly tied to veterinary use of fluoroquinolones.
This and other reports from this laboratory were what the industry viewed as the trigger which set CVM on their current path to propose sweeping changes to the regulatory process.
While we viewed this information as important regarding an emerging a food borne threat, we did not believe that the information was sufficient to cause such a significant disruption to the current approval process for veterinary drugs.
First, the use of the term "resistant" has been used by the authors not to describe clinical resistance, but rather a shift in susceptibility. They have chosen lower breakpoints than the standards set by the National Committee for Clinical Laboratory Standards (NCCLS) and the British Society for Antimicrobial Chemotherapy. What have been reported as "resistant" isolates are in reality clinically susceptible according to the NCCLS and BSAC guidelines.
Second, as far as we know, there has not been a documented case of a human fluoroquinolone treatment failure in the UK because of DT104 as a result of the treatment of animals.
Third, reports from that same laboratory over the last two years demonstrate a marked decline in the incidence of Salmonella typhimurium DT104 and no clinical resistance to the fluoroquinolones has yet emerged. At the same time, the incidence of DT104 with increased MICs to fluoroquinolones has really not changed.
Another study concerns fluoroquinolone resistance levels in Campylobacter species in poultry in the Netherlands published in 1991. This information was considered by the 1994 FDA Joint Advisory Committee prior to it recommending that the fluoroquinolones were approvable for therapeutic use in food animals with certain restrictions.
The Advisory Committee did not consider the Netherlands experience adequate evidence establishing a public health risk to preclude the approval of quinolone animal drugs in poultry. For one thing, a high level of resistance was already present in Campylobacter prior to the introduction of fluoroquinolones for use in poultry.
The study from Spain was mentioned earlier, where increases in resistant strains of Campylobacter species were, in fact, observed, however, Spain is a country where manufacturing, distribution, and sales of relatively low quality generics do abound, and other veterinary and human pharmaceuticals are generally less controlled.
In particular, these products tend to be more readily available, as was mentioned, for human and animal use without prescription, in contrast to the limited and veterinary controlled uses in the United States. It is important that we make that difference.
This report also failed to demonstrate that there was a direct link between the use of fluoroquinolone in animals and the actual development of resistance that was determined in people.
Another reference from the Minnesota Department of Health has also been referred to here today. That data is yet to be published, so we really don't know exactly what it says, but we have heard at various meetings pieces of it.
From the information we know about, only a very small percentage of human clinical cases were associated with the fluoroquinolone resistant Campylobacter, and the majority of these were attributed to foreign travel.
The same author has reported that fluoroquinolone resistant Campylobacter has been increasing in human isolates since 1991 in Minnesota, and that is four years prior to the approval of any fluoroquinolone in food producing animals.
Now, the document also points out concern for development of antibiotic resistance in non pathogenic enteric bacteria, which may under certain circumstances be pathogenic. References are appended from several studies in Europe suggesting a link between vancomycin resistant enterococci and glycopeptide use in animal feeds. We have heard a discussion about that this afternoon.
These references represent a significant research effort in Europe to incriminate the use of antimicrobial growth promoters as being responsible for transferring resistance to humans.
I would comment that these and other studies have been considered by the Scientific Committee on Animal Nutrition, an advisory body to the European Union Commission.
They have reviewed the situation with several drugs, avoparcin, virginiamycin, tylosin, and spiramycin, all the drugs that the European Union has decided to ban. In every case, their conclusions have been that the data falls short of being able to conclude that the use of these drugs in animal feed represent a significant public health risk. However, as we know, the European Union moved ahead with their ban.
Now, there is no question that common resistant isolates or resistance determinants can be found in humans and animals as a result of antibiotic use. Clearly, animals and humans can exchange bacteria carrying these properties. I think we have seen evidence for that. However, the cited evidence in the framework document, in our view, simply does not rise to a level which justifies the extreme measures being proposed here by CVM. This does not mean that we shouldn't take safeguards, and we will try to discuss what we think is our approach to the problem later in this presentation.
Now, let me talk a few minutes about some of the specifics on the proposal, so you can get our views of it.
With regard to categorization, the agency is proposing that the human health impact will be evaluated on two factors: one, the importance to human medicine; and two, the potential human exposure. That was discussed earlier by Dr. Sundlof.
Based on this evaluation, FDA proposes placing the antimicrobials into three categories based on their value to human medicine and their exposure.
Now, AHI shares the concern for preserving the usefulness of antimicrobial drugs for treatment of human infections, while at the same time balancing the need to assure the availability of needed antimicrobials in food animals.
However, we believe the plan proposed by CVM will likely assure that development of important new antimicrobials for food producing animals may not even be attempted, as Dr. Fox alluded to.
A significant problem with establishing pre approval and post approval requirements based on the categorization is a dynamic new process by which pathogens emerge and new antimicrobials are discovered and developed.
Because new drugs in discovery require 10 or more years to develop, it won't be possible at the time of discovery to really project the importance of a new antimicrobial to human medicine.
That, of course, will be dependent on diseases of importance to humans and availability of other effective drugs at the time of expected commercialization of the new antimicrobial.
Because virtually any class of antimicrobial that has the potential benefit for animals will have similar benefits for human medicine, it is really difficult to imagine that any innovative antimicrobial would be developed for animal use without really having to meet the criteria of Category I, and we recognize there are several categories, but to us it appears that most drugs are going to fall into Category I, and this is obviously going to lead to a reluctance by companies to invest in their development.
The result, of course, will be more reliance on the older products, and hence, more resistance selection for those older products.
Now, some might suggest that animal health companies should just develop drugs for animal use, and avoid anything related to human medicine. Well, as I said before, this is rather difficult because any drugs that have a potential for treating human disease will probably have applications in veterinary medicine, and, in fact, most animal health companies share their discovery research with their human counterparts.
The economics of trying to do discovery research for animal drugs only simply doesn't make sense and certainly can't be justified economically.
Further, what might not be important today for medical uses might become important in the future. So, it is a very difficult balancing act how do you determine what is important to human medicine today, so that you have that vision for the future.
CMV also talks about exposure scenarios, and AHI certainly agrees that potential exposure of humans to resistant organisms is important to consider. In fact, we believe it is the primary factor to consider.
FDA states, and AHI concurs, that antimicrobial resistance transfer is determined by a complex chain of events. The proposal lists many factors that should be considered when classifying potential exposure.
These include attributes, product use, and potential human contact. Although food processing is mentioned, the emphasis is clearly on the attributes of the drug and how the product is used on the farm.
The industry sees a problem with this. The number of animals treated, for example, has little relationship to actual human exposure to food borne bacteria.
Clearly, the most critical factors in determining potential exposure take place after the animal or food products, in the case of milk, leaves the farm. For example, consider the use of antimicrobials in dairy calves. Exposure to pathogens, whether they be susceptibility or resistant, is eliminated with pasteurization. The risk essentially is zero assuming there are no failures in the pasteurization process.
So, drug attributes, product use, potential human contact, manure management practices, a lot of these factors are essentially non factors.
Of course, we have a different situation with meat and eggs. These products are not pathogen free. However, we are all aware of steps that are being taken, such as HACCP, steam sterilization, irradiation, that should have a major effect on reducing food borne pathogens from a number of animal sources.
AHI doesn't believe that this important aspect relating to exposure has really been given adequate consideration by CVM in the development of their proposal.
Let me comment a moment on pre approval studies. The framework proposes that pre approval studies would be necessary for all Category I and II to assess the rate and extent of resistance development in enteric bacteria.
The document also talks about resistance thresholds and monitoring thresholds. For Category I, the agency says it may be possible to establish a level of resistance that will not cause a significant transfer to human pathogens.
However, lacking that data, the agency would consider any level of resistance change to be a cause for the drug not being shown to be safe. In other words, the drug sponsors must demonstrate by pre approval studies what level of resistance is safe prior to approval.
We believe the concept here proposes a standard that simply can't be met. Aside from the fact that the document is unclear as to whether these thresholds are based on susceptibility shifts or clinical resistance, the Center is acknowledging that in many cases it won't even be possible to define a safe level of resistance.
Since there is very little correlation between in vitro susceptibility of enteric bacteria from food animals and impacts on human health, there is little likelihood that you could ever set a safe level of resistance. Therefore, the agency, we believe, is proposing a rather prohibitive standard given the fact that resistance development is a natural response by bacteria.
Furthermore, it appears that CVM may be using a similar concept and I think others have commented on this to the way animal drug residues are handled, but there are important differences which make that an unworkable approach. I think Lyle Vogel commented on that.
At least with drug residues, we have assays, we have safety factors, statistics can be applied. The scientific basis and protocols for establishing resistance standards that are similar to drug residue tolerances simply haven't been developed. There isn't a long history of toxicological research that has gone into antibiotic resistance. It simply doesn't work to really quantify resistance by the methods used to establish residue tolerances.
Pathogen load. We have some concerns about pathogen load. FDA suggests that this is necessary to determine the time required for the pathogen load to decrease following treatment. We question the basis for this requirement.
Implicit in the requirement for pathogen load studies is the assumption that quantitative viable counts of pathogens, above a baseline normal, will present a greater risk to public health.
We are not really aware of evidence that correlates increased on farm gut concentration or prevalence of food borne pathogens to increased human disease from those pathogens, nor are we aware of data which indicate that shedding of gram negative bacteria, which are sensitive to a drug under test conditions and that would be the case with any new products should even be of concern with broad spectrum antimicrobials.
I think we heard this morning the use of a resistant strain. Well, that seems to be imprudent to develop resistant strains just to do studies.
There are a number of inherent difficulties that can be pointed out if one attempts to acquire the information, and I think it was already mentioned that there are some studies in swine, I won't go into that, but these on farm studies that USDA has collected have shown a multitude of factors that contribute to pathogen shedding, and transportation is certainly one of those.
Establishing a relationship, a clear relationship between pathogen load and the use of the drug, we think is a very difficult thing to do, confounded by many factors.
Let me move to post approval studies. It is clear that FDA believes that on farm studies to monitor antimicrobial resistance development will be necessary for all Category I and Category II drugs, again, to ensure that thresholds are not exceeded.
The proposal would have drug companies collect such data on a drug by drug basis to establish and monitor these farms to meet the established monitoring and resistance thresholds, so that intervention and mitigation strategies could be investigated and initiated in a timely fashion.
AHI has serious concerns with this concept. We don't believe that on farm isolation and susceptibility testing of food borne bacteria, in particular pathogenic organisms, represents the best or most efficient location for assessing exposure.
Because of the relatively low prevalence of pathogens, numerous animals would need to be sampled in order to gather meaningful statistically valid data upon which to determine changes in susceptibility.
Now, in order to get around these problems, CVM has suggested that surrogate organisms might be used as sentinels for pathogen changes. We are concerned that the use of a surrogate removes the relevance of the results even further from what we are trying to accomplish, that is, to assure food safety.
The framework lays out FDA's belief that it would be appropriate to evaluate mitigation measures. Now, we are certainly interested in determining mitigation measures that could be used to decrease the rate and extent of resistance development. The information would be helpful to our companies in prolonging the effectiveness of antimicrobials. However, we don't see how such studies can really be justified as part of the approval process.
Information from these studies should be used in the judicious use initiative, and this is an area where industry, the veterinary profession, and government should work together, but we don't think it belongs in the drug approval process.
Now, as you will hear in a few minutes, we believe the best early warning system to monitor for changes is not on the farm, but in the slaughterhouse and closer to the consumer of meat and poultry. Further, we view testing for food safety purposes to a federal government responsibility as it is with other food borne hazards, such as animal drug residues and pesticides.
The costs of on farm testing should not be underestimated, or the logistics of even trying to collect representative data to determine if a pre determined quantitative threshold has been exceeded. Estimates run more than a million dollars per drug per year even if studies could be adequately designed and conducted, and that is probably an underestimate.
The scope of testing that CVM has in mind, we believe might be beyond even what the federal government is capable of doing in the surveys that FSIS and APHIS have conducted over the years.
Thresholds. It is not clear in the document what is meant by a "threshold," whether it's a resistant or monitoring threshold and how the two may differ. We are assuming a resistance threshold might be a higher value than established for monitoring. If that is the case, then, we have complicated an already difficult process and added yet another set of assumptions to the approach. We have not only one threshold, but multiple thresholds. It is getting very difficult.
The use of in vitro susceptibility data as a regulatory tool, I believe has many drawbacks. Now, susceptibility testing is very valuable for evaluating trends and useful as an indicator for selecting therapeutics, but it is a measure of in vitro activity and in no way assures therapeutic outcome. It's a laboratory test. When in vitro susceptibility testing is used as a monitoring tool, we have been told by experts in the field that several years of data are really necessary to establish trends before you could tell whether something is occurring, and although shifts may be detected in the short term, more time is needed to confirm these trends.
The Salmonella DT104 situation in the UK, that I have mentioned earlier, is a good example of that, whereas, shifts initially were seen, and they seem to be leveling off.
With that, I think I will close and turn to my partner, Mr. Mathews, but as you can see, FDA's proposed framework for regulating antimicrobials, AHI does not believe can be practically implemented.
In closing, I want to urge you in your role as advisers to the Center for Veterinary Medicine to request that the agency reconsider its proposal for a change in the regulation of animal drugs as they have suggested.
|Israel journal of veterinary medicine vol. 63 No. 1 2008||Ranch hand advisory committee|
|Advisory committee on immunization practices||Medical Devices Advisory Committee|
|External Advisory Committee on Cities and Communities||Wildlife Diversity Policy Advisory Committee|
|National Vaccine Advisory Committee (nvac)||Peer reviewed by the Arizona Department of Commerce Economic Research Advisory Committee|
|College of veterinary medicine north carolina state university m. Christine McGahan, Professor and Department Head||College of veterinary medicine north carolina state university m. Christine McGahan, Professor and Department Head|