• 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

The Animal Health Industry by Thomas Shryock, Ph.D., Animal Health Institute

DR. SHRYOCK: Thank you Tom and Pat for the opportunity to be here this afternoon and kind of on the cleanup side of things, so in the late afternoon.

It is my pleasure to represent the animal health industry based at Animal Health Institute in Washington, D.C. The comments that I will make today are reflective from that perspective.


Real quick, we have five simple objectives here. As you can see on the screen of what will be covered. So we will start in more from the perspective on these objectives, not to go over them one by one, but just to include them for reference in the slide deck that will be available to you post-meeting.

As you can see, these are the ones that were listed from the 2007 FDA Science Board Review. And then we also have those same objectives reworded a little but perhaps as posted on the NARMS website from June 2009.

So they are basically the same and we have covered those already so I won’t spend additional time this afternoon going over those. I think we are all very well familiar with them. 


What I would like to share with you is the perception of industry needs that relate back to those NARMS objectives and the NARMS value to the industry. There is great opportunity for supporting responsible use. We have heard the number of different producer guidelines and veterinarian guidelines that are already here.

The idea with NARMS data is that we can measure to manage. What we need to be able to manage, obviously is foodborne antimicrobial resistant pathogens. One of the challenges with having data on foodborne pathogens for veterinarians is that they really make their use decisions on antibiotics based on target animal pathogens, those that cause disease in the animals, whether it is bovine respiratory disease, swine enteric disease, or what have you.

So they are looking at an entirely different set of pathogens which are evaluated in veterinary clinical diagnostic labs using CLSI methodology and perfuse clinical breakpoints we trust to make those decisions on use, rather than referring to the NARMS data when they are in the fields faced with a use decision.

So responsible use is supported, but there are some limitations into how far that could be extended.

From a regulatory prospective and this hits the manufacturers of animal health antibiotics, per Guidance 152 which is the risk assessment document that all new animal drug applications have to follow and for which some folks are recommending it be used for current products on the market.

These antibiotics are subject to NARMS testing if you read the document. However, NARMS data is only “supportive” for new claim applications at CVM. In some cases, that data may be usable. In other cases, perhaps not.  I think it is on a variable basis. So there is some limited utility there.

One of the objectives of NARMS is that it was a source of isolates which some of the sponsors or perhaps other researchers might want to acquire for their own study purposes. In this particular line, it would be for conducting susceptibility studies on new animal drugs. But the process for obtaining those isolates is not really clear and could use some better definition.

Finally, from a food safety and risk assessment use, again where we are looking to compare say a baseline versus change, do we have the right data out there. Is that something that could be improved?

So to summarize all this from an industry prospective, the value to the animal health industry is somewhat limited. Does that mean that the program is not supported? Not at all. In fact, there is great support for it.


Let’s look at some of the data here. We have heard a lot of interpretations of the data from a variety of folks over the last day or so, but what the folks around the table at the Animal Health Institute and Scientific Advisory Board wanted to do was just take this data at face value and see what the data looked like, looking for trends as the data was intended to be used for.

We are not aware of any similar nongovernmental-type of review that is out there. This certainly is not published as a review here, these next few slides. But I just thought it might be interesting to take a quick look at some of the critically important antibiotics as defined in guidance 152 looking at both Salmonella and Campylobacter in particular.


So rather than go over all these bits and pieces of data since it is so late in the afternoon, they are all listed here, but the trends that I think we can start to look at and which we have already see for some of the nontyphoidal Salmonella continue to show very low frequencies of resistance among human isolates as well as slaughter and retail sources as illustrated here.

So that is one key thing that oftentimes we talk about how much resistance. Well here is some data that we need to take into account and compare these non-typhoidal Salmonella which may be obtained from a variety of attributable sources to those Salmonella Typhi which were also discussed this afternoon, just to look at those resistance prevalence rates which are in the same ballpark. So for what that is worth.


Looking at Campylobacter jejuni and the fluoroquinolones, again we have just heard some comments relative to some of the poultry uses of a product. Here is the data, looking at chicken slaughter isolates. In 2004, the product would have been available versus 2007 after it has been discontinued. There are the numbers.

Human isolates had a little bit of an increase, an upward swing in 2007. However, the only attribution there, where are those human isolates sourced? It is kind of difficult to know off hand. 

So for what it is worth, another temporal observation.


Cephalosporins, this would be particular in regard to the third generation cephalosporins, not the first and second. Again, looking at the non-typhoidal Salmonella human isolates are pretty low percentages. It looks like it is fairly stable, maybe trending down a shade. But the interesting thing here is retail ground beef and pork chops, nothing recovered.

So oftentimes that kind of gets lost in the scheme of things. So that is a very important break in a continuum of the food chain.

Ceftriaxone resistance kind of hand-in-glove with the ceftiofur resistance patterns here as well. Just for contrast purposes, the Typhi very low overall resistance prevalence, a little bit lower, perhaps, than what was reported for the NTS isolates.


Macrolides to round out the critically important antibiotic classes. Again, a trend line that is very stable over a period of time. Looking pretty much here at the human isolates, very low numbers that are recovered so a percentage at three to eight percent per year represents three to eight isolates.

So there is some interesting data there, just to look at trends as we go forwards. And if we look at the numbers of classes resistant and all that, you can see where that comes out here.


As far as multidrug resistant Salmonella and Campylobacter, isolates from humans, again, just some general observations that came out. 81 percent of the Salmonella in 2007 were pansusceptible. I think we are often focusing on this 19 percent resistant or even a small subset that has got more resistance.

The bottom line is nobody really wants Salmonella in their food, but yet we tend to focus in on that subset that is MDR and even that fraction that is there that is pansusceptible, is a little bit of concern and needs some attention.

So we can see some of these declining resistance trends as we move down here through the data. So again, temporal trends for what it is worth.


So what can we really conclude from the data? Well, you know, again, I don’t think there is a whole lot of numbers here to look at. It is a really interesting just temporal relationship that we can see here, not only for the fluoroquinolones, but the macrolides, as well as for the cephalosporins.

I might point out that in your slide pack, this particular slide I amended some of the numbers here as I detected some errors in that, so you may want to make those appropriate corrections.


So just to conclude on that at a very, very high level, just looking at those numbers. This is not a risk assessment or anything, but the data would suggest that the transfer of animal isolates including those that are resistant to any of those three critically important antibiotic classes onto meat and then to humans would be an infrequent occurrence. 

Then as we talk about MDR isolates of Salmonella and Campylobacter, those are a small percentage of the overall isolates that are recovered.


So let’s talk about some desired NARMS enhancements. That was really what I was charged to provide some perspectives on when I was invited to represent the industry prospectus. So as I mentioned, AHI does support the NARMS program but believes it can be improved to increase its usefulness.

A lot of the FDA Science Board recommendations that had been made would be generally endorsed. We have not heard a whole lot about those. I do acknowledge and do appreciate Pat’s willingness to share strategic goals in some of those revisions and I think those are very much in line with these FDA Science Board recommendations.


So again, I will not go through all of these. There are many that are included as excerpts from the full report that are in the printed text before you and it will be on the slide deck available post meeting.


The one piece that I want to focus on a little bit more is the strategy for sampling. One phrase that really jumped out to me as I was reading this in general terms is that the subcommittee believed that resistant data must be able to withstand legal and regulatory scrutiny and challenges. 

That caught my eye because that really means that if we are going to be studying policy and making decisions on approval or removal of antimicrobial products, it has got to be quality data. It has got to be a quality sampling system. It has got to be quality everything. It is just that important.

The downside and I think you can see this here is if it is not done, then there is trouble. And I don’t think anybody wants to go there.


So the Science Board had made had made these following recommendations. Here are the ones on retail meat. You know, there are challenges. It is a tough, tough thing to go out and collect all those. There is expense, there are other issues. Pat highlighted a few of those challenges in some of his remarks the other day.


For the USDA, there is probably a need to reconsider the sampling strategy, getting the isolates from FSIS that support HACCP. Maybe there is a better way to do that, a more specifically NARMS-designed kind of program.

The CAHFSE program received endorsement from the Science Board. That offers an excellent opportunity to really get to the point where you can have drug use information correlated with clinical outcomes and all sorts of context information. It would be extremely valuable to support responsible use.


So a continuum is what would really be desirable. We talk about this in risk assessment and I use these terms release exposure consequence. It is what is in guidance 152. It is what is in the OIE terrestrial code risk assessment. 

All that is meaning is that we are going from food animals to retail meat, the food, and ultimately into patients. We won’t have clinical outcome data here in the patient population, but at least we can start to fill in or identify data gaps that need to be filled in along a risk assessment continuum.

Obviously you can see all these wonderful things that we can support with quality data and a well-designed system.


So that is kind of the desired continuum. It gives context to the information rather than looking at bugs over here and finding a few over here that might have a similar profile and drawing a straight line. There are a lot of intermediate steps in there that a risk assessment could be very useful to help identify and provide further information on.


Another way of looking at this, although it will not show up very well here on the screen, is to look at how the NARMS testing matrix of meat and live animal and human data matches up with the bugs of interest. Right now you can track Salmonella. It is sampled the whole way across. But as we go for Campylobacter, E. coli, and Entercoccus, there are some gaps in here along the way.

Is it worthwhile to try to fill those in? I will leave that as an open question. We know Shigella and the Salmonella Typhi down here in the orange section. It does not really make a lot of sense to look for those in anything to do with meats or in animals, so of course that is going to be not relevant on that.


So it is kind of disconnected. If you wanted to fill in all those matrix boxes, we can’t get it done exactly. So right now, how do we -- if we fill in that box to meet that NARMS continuum need. 

We have gaps in terms of the pork chops and ground beef. There are not enough Campylobacter or Salmonella coming back out to really do much with those.

Richard Reed-Smith  had acknowledged in the CIPARS program they have stopped and they will collect what they collect, but they are not going to go out looking for those bugs because the sample requirements would be too great.

So does that tell us something about what we might want to consider to conserve resources within the NARMS program?


Which leads then to prioritizing. Does it make sense to expend resources looking for bugs in products that are very difficult to find in the first place?

What about the commensals? Should we still continue to look for those? Sure they are resistance reservoirs in theory, but is that a good use of resources when they do not cause foodborne disease per se.

Looking at O157 for antibiotic resistance when it is not treatable with antibiotics; Is that a good use of NARMS resources? Why not just do that as a one off * in another manner? Do we have the right bugs and drugs combinations? Should we already by considering those and changing them out as has already been done? Bacitracin and Enterococci discontinued; it doesn’t make sense. For example, Lincomycin and Entercocci, is that a really relevant combination? Do we need three aminoglycosides versus Salmonella? Just some open questions for consideration on prioritization, trying to make sure we are applying the resources in the right way.


So I guess the bottom line is does it make sense to continue to do things the way they have always been done? Wouldn’t it make sense really to put a NARMS-specific sampling strategy and plan for that animal, carcass, or meat continuum for foodborne pathogens in place? We will assume the human pathogen isolations will continue going into the future.


You know, NARMS was a good start. It is a good program as it is, but there are a whole lot of things that are going on right now that we really need the quality strategically designed representative sampling strategy to address and to meet these particular goals.


So that implies that NARMS will need to reconfigure. Very consistent with what the FDA Science Board recommendations put on the table.

Another piece to this is having adequate funding for the program. It seems like if this is an important national objective to have a measurement system in place to help guide policy, responsible use as well as all the other things so listed, that the resources should be made available. Rather than investigators spending time trying to secure funding for research on a competitive basis, why can’t we secure dedicated funding for this particular initiative?


So the sampling strategy piece, I will come back to that and just leave you with this thought. Again, the resistance data has to withstand legal and regulatory scrutiny and challenge. That is the Science Board wording, not mine. Of course, then the consequences will limit the value of the NARMS findings in the long term.

So that is the conclusion of what I had to say at this point in time. I leave you with those suggestions. I do not have specific recommendations on how to improve the program. Why? Because I think this is a multi-stakeholder issue and all of us are smarter than just me.

So with that, I will be glad to turn it back to the moderators. Thank you very much.


DR. FRYE: Thank you Tom.