HHS Listening Sessions on AMR | February 2020 | Day 2, Part 1

By Adem Lewis / in , , , , , , , , , , , , , , , , , , , , /

>>Jomana Musmar: All right. Thank you, and welcome to day two of our AMR
listening sessions here, hosted by HHS. We’re glad to have all of our presenters join
us today. For those of you joining us again for day
two, we appreciate you staying. We thank our participants that are here on
an individual basis — subject matter experts from all walks of life — and our agencies,
from DOD, USDA, and HHS. Again, the goal of this meeting, as a continuation
from yesterday, is to learn about ways to improve the programs and policies related
to antibiotic resistance that would contribute to the overall health of our country. We will not be asking this group again to
vote on any matters, or make any recommendations, or reach a consensus view during the meeting
today. The purpose of today, again, is to discuss
the One Health issue surrounding AMR with an exchange of ideas, for informational purposes,
only. So, as such, we’ll be soliciting views on
an individual basis from all folks that are sitting around the table today. So, just as yesterday, I’d like to go through
roll call, just to acknowledge everyone that’s with us. We’ll start with Marty Blaser.>>Martin Blaser: Here.>>Jomana Musmar: Mike Apley?>>Michael Apley: Here.>>Jomana Musmar: Stephanie Black?>>Stephanie Black: Here.>>Jomana Musmar: Helen Boucher?>>Helen Boucher: Here.>>Jomana Musmar: Sara Cosgrove, on the phone?>>Sara Cosgrove: Here.>>Jomana Musmar: Thank you. Paula Cray?>>Paula Cray: Here.>>Jomana Musmar: Christine Ginocchio?>>Christine Ginocchio: Here.>>Jomana Musmar: Locke Karriker?>>Locke Karriker: Here.>>Jomana Musmar: Lonnie King?>>Lonnie King: Here.>>Jomana Musmar: Kent Kester?>>Kent Kester: Here.>>Jomana Musmar: Elaine Larson.>>Elaine Larson: Here.>>Jomana Musmar: Ramanan Laxminarayan?>>Ramanan Laxminarayan: Here.>>Jomana Musmar: Armando Nahum?>>Armando Nahum: Here.>>Jomana Musmar: Paul Plummer.>>Paul Plummer: Here.>>Jomana Musmar: David White.>>David White: Here.>>Jomana Musmar: Denise Toney?>>Denise Toney: Here.>>Jomana Musmar: Tiffany Lee?>>Tiffany Lee: Here.>>Jomana Musmar: Kathy Talkington won’t be
able to join us. And on behalf of CDC, Rima Khabbaz?>>Rima Khabbaz: Here.>>Jomana Musmar: Dennis Dixon, NIH?>>Dennis Dixon: Here.>>Jomana Musmar: Chris Houchens, BARDA?>>Christopher Houchens: Here.>>Jomana Musmar: CMS, on the line? Bill Flynn, FDA?>>William Flynn: Here.>>Jomana Musmar: Thank you. OGA, Larry Kerr or Lynn Filpi? Paige Waterman, DOD? USDA, Chelsea Cheveley [spelled phonetically]? USDA, Jeff Silverstein?>>Roxann Motroni: Roxann Motroni, here for
Jeff Silverstein.>>Jomana Musmar: Thank you, Roxann. And USDA, Emilio Esteban.>>Emilio Esteban: Here.>>Jomana Musmar: Thank you. And the same as yesterday — same housekeeping
rules. All participants, please identify yourself
before speaking into the microphone. It’s important that this is on because this
meeting is being streamed live. It’s being transcribed, and we want to ensure
that the statements are attributed to the individuals making them. Please make sure to turn off your speakers
when you are done, and try to get as close as you can so that the folks that are just
transcribing this and the web stream can pick up your voice. And let’s see, that’s it. Thank you so much for joining us today, and
I’ll turn it over to Dr. Blaser.>>Martin Blaser: Thank you, Jomana. I’d like to join in welcoming everyone, with
Dr. Lonnie King. We want to welcome everyone back to day two. Yesterday was an exciting day. We heard about many promising technologies. We’ll be hearing more about others today,
and expanding our reach, and looking at some new areas that we hadn’t discussed before,
like the microbiome, and like aquaculture. First, we’ll hear an update from the CDC on
its recently released antibiotic resistance threat report. I don’t have to remind everyone that we are
in the times of coronavirus, starting now. And unfortunately, we could predict that with
a lot of sick people, there will be more manifestation of AMR, because there will be more people
in ICUs, and more lines. And so, this makes our work especially important. So, we turn to Dr. Michael Craig from CDC. Thank you.>>Michael Craig: Thanks to the council. Really delighted, here, to present on CDC’s
A.R. threats report, and talk about all the work that we’ve been doing. Before I begin, I want to just highlight that
in the 2019 threats report, which came out in November — and it’s an update to our earlier
2013 threats report — we have a foreword and a dedication to all the patients and families
who have been affected by antibiotic resistance and C. diff. I want to commend the committee for all of
their work yesterday to highlight some of those patients’ stories. It’s always important for us to be grounded
in the human impact, the daily impact, the community impact, the family impact of these
infections and what they can do. And it’s always good for us to be reminded
and go back to that, because at the end of the day, that’s what this is about. What I also want to just highlight and commend
everybody at CDC and our public health departments for, the work that went into this report. It’s obviously not the report of a single
person, or even a single team. This is really an — this report is one of
the — those reports that really pulls together information all across CDC. And hundreds of people really contributed
to the success of this report, and we’re very happy, and I’m very humbled to be able to
present all of the information to you today. Next slide. So, I think folks know this was the numbers
that we had in 2013. And I think a couple words here — these are
the numbers that people sort of became accustomed to, and we released for the first time. It was the first time that CDC provided a
national estimate of all of the threats that we saw for — on antibiotic resistance, and
pulled them together, in an estimate. This was in 2013, which means that it was
before we had a national strategy, before we had a national action plan, before we had
a PACCARB. And that’s an — important to note, because
it was the best data that we had available at the time, but it wasn’t the best data. And we knew, and we provided an estimate with
what we had available to us, but it was really — there were a number of limitations with
the data, and we described those limitations in full detail, and we always told folks that
that was a conservative estimate, based upon what we had available to us. Next slide. And so, I would note that when we took the
opportunity to update our threats report, we took a lot of care into consideration of,
how could we improve upon the data, and how could we provide new information to not only
give a better contemporary estimate, but how could we also look back and actually go back
in time, and provide a better estimate of what the actual burden was when we released
the first threats report? So, very pleased to highlight the fact that
we did this. And it was one of those where we sort of pulled
together new data — and would highlight here that the data that we pulled together was
a very innovative pulling together of three different electronic health record systems
from three different EHR vendors, that really magnified and increased the amount of analytical
rigor that we had available to us, to come up with not just a contemporary estimate to
— but to go back in time, provide an updated 2013 recalculation, and then, provide trend
information for many of those pathogens. And so, as you can see here, we provide sort
of a snapshot of what the number actually was in 2013, with the 20 — the 2.6 million
illnesses, and then, ultimately, the 44,000 deaths. So, the death number was a significant underestimate,
as we said at the time. And as we’ve said since then, it was something
that — we always considered the 23,000 and the 2 million to be conservative, and this
really gives us a snapshot of how conservative the estimate was. Next slide. And then, that brings us to present day. These are the more contemporary estimates
that we released and provide a snapshot of where we are. And it’s important to note a couple things:
like the 2013 threat report, the 2019 threat report’s wallet included C. diff. We did separate the numbers there somewhat,
because C. diff, as folks know, is slightly different than the rest of the A.R infections,
but they are related enough that we want to include them in the report. So, you see the over 2.8 million infections,
and the nearly 36,000 deaths from the A.R. pathogens, plus the nearly quarter of a million
cases — hospitalized cases of C. diff, and the over 12,000 deaths. So, combined, all of the pathogens in the
threats report — we’re actually talking about over 3 million infections, and nearly 50,000
— over 48,000 deaths. And you know, the — well, we would sort of
note there — there’s a couple of things we would note, is that because of the recalculation
that we did, our top line message on this is that the problem with antibiotic resistance
in the United States is worse than we previously estimated. That — as you saw in the previous slide,
the 44,000 number is significantly higher than we previously estimated with 23 — from
the 23,000 we estimated in 2013. That being said, another important theme that
we want to note, and will highlight — next slide — is that we have actually seen prevention,
and we’ve seen a trend of prevention of A.R. infections and A.R. deaths since 2013. And our trends really do show that for especially
the healthcare-associated resistant infections, we have made significant progress in the United
States at preventing those. As you can see from this slide, and as we
just want to note — is that prevention really does work. And I was really heartened by a lot of the
conversation yesterday of the importance of reinforcing prevention messages, and I think
this really shows the impact and the significant work not just by CDC, but by state health
departments, by CMS, by really the — all of the U.S. government, in terms of focusing
on prevention. And as you can see, we have, overall, that
decline from 44,000 deaths to the 36,000 deaths — is the 18 percent. But if you actually look just at the specific
hospital-associated ones, that’s really what’s driving the change here. And then, at the bottom, you can see some
of the very pathogen-specific — some incredibly significant numbers there. We also would just note that we highlight
CRE as being stable, and we say that that’s a success. And we highlight that because in a lot of
countries, we have seen the growth of CRE and the concern about CRE, and we do consider
it a success that we’ve actually maintained that level in our country over the past few
years. Next slide. But also want to highlight — is the other
side of the coin — is that we have the same message that this is still worse than we previously
estimated. And in fact, as was noted yesterday, a lot
of the infections on the community side are increasing. Would highlight a couple here — the urgent
threat of drug-resistant gonorrhea continues to increase very significantly — and then,
the ESBL-producing Enterobacteriaceae increase that was highlighted somewhat in the discussion
about vaccines yesterday — we’re seeing significant increases here. These are important because these are really
outpatient UTIs. These are the urinary tract infections that
disproportionately affect women. And these are the infections where previously,
you might have gotten an outpatient prescription, and it be remedied. But now, because of resistance, some of these
infections — in fact, many of these infections — are requiring hospitalizations with stronger
and more complicated treatment. Next slide. As we did with the first threats report, we
categorized all of the threats into three different threat levels: urgent, serious,
and concerning. And the next slide. As you can see here, with that ranking, we
did have some updates, and we’ll just highlight those briefly for you here. Two of them, we had changes in our highest
category, the urgent category. One — not surprisingly, given all of the
conversations that you’ve heard, as well as what’s been in the news — is Candida auris,
the drug-resistant yeast. Was not listed in the 2013 threat report because
nobody in the world knew about it in 2013. And I think it is — really highlights the
fact and the challenge that we face of emerging pathogens that we don’t know about, and then,
circumnavigate the globe, and can cause very significant illness, including death. And so, that was something that we didn’t
know about in 2013, and now that we do, and have really worked globally to try and address
it and to work with, especially, healthcare facilities and long-term care facilities,
where we see many of these cases — to identify it and respond to it. It went to our highest category, in the urgent
category. And then, the other one we’d highlight here
for you in the urgent category was Acinetobacter. Last time, we had it listed as multi-drug-resistant
Acinetobacter. For a number of reasons, our subject matter
experts wanted to focus, instead, on the carbapenem resistant Acinetobacter, because that is the
more concerning form of resistance that we’re worried about. And when we looked at just the carbapenem
resistant Acinetobacter, we — and we compared it to the other urgent threats and the threats
in the report, we noted that it really did — it was comparable to CRE in many ways. And so, we moved its threat level to urgent. And then, the other one we just briefly noted
is that VRSA was in the previous report as a concerning pathogen. We actually removed this as a threat and combined
it with MRSA, and that’s largely because we have not seen cases of VRSA. And in fact, it’s not — doesn’t have the
same level of concern that we once had for it. Next slide. The threats report also — this time, we included
a new category, and this was a watch list. This is something where we did not provide
estimates of infections or deaths, largely because these were pathogens that we’re sort
of paying attention to, that we think, ultimately, might be categorized in a future report. We think they’re important and really, we
think that they need to be given special attention, and more attention needs to be brought to
them. And really, they’re not well understood, or
they’re very infrequent in the U.S. Can highlight some of those, briefly. As has been discussed here previously from
some of CDC’s fungal experts, we have growing concern about azole-resistant Aspergillus
fumigatus. This is a pathogen that really underlines
the importance of the work on One Health, because we see azoles that are used for spraying
crops, in the environment. And we are now seeing that the spraying of
those crops with azoles is leading to types of resistance that are affecting humans. This has been well-documented in Europe. We’ve had some cases that we’ve reported on
in the MMWR in the CDC, and it’s something that we’re looking for, and trying to identify
what is going on with this, and what, ultimately, could be done to mitigate some of this. The other pathogens we’d highlight there,
briefly — M. genitalium is a sexually transmitted infection. And going hand-in-hand with our concern about
drug-resistant gonorrhea, this is, again, something that’s poorly understood, and we
don’t fully understand the scope of it globally, let alone in the United States. And we’re trying to get a better handle on
not just the prevalence of the infection, itself, but the issues of drug resistance
that go along with it. There are not good tests for M. genitalium,
which is some of the limitations for understanding it. The other piece we’d highlight is pertussis. This is, of course, a vaccine-preventable
disease. The challenge that we see, and the reason
why we’re highlighting this, is that in parts of the world, we are seeing resistant forms
of pertussis. Would note that with the vaccine, it will
prevent the resistant forms. But as with anything, when you start to see
something where we have a vaccine-preventable disease, and it’s spread, and we see resistant
versions of it, it is a growing concern. We have not identified any drug-resistant
strains of it in the U.S. to date, but we note they have been identified in other parts
of the world. Next slide. The other thing I want to highlight with the
threats report is our very intentional messaging. I hope folks here have had a chance to look
at it. I will say that — and I hope — we aspire
to have it be the most One Health report the agency has ever put out. Some folks have told us that. I hope you guys feel the same way. But we did take a lot of interest. We had a lot of conversations with a lot of
different groups across a lot of different industries on how we could improve our messaging,
on how we could improve our engagement. As we say in the foreword of the report, the
intention is not to blame anyone, but is to really reflect the nature of the One Health
problem that faces all of us, and really call all of us to collective action. And the — as you can see in the next few
slides, these are some of the things that we’ve highlighted, to try and highlight the
importance of the One Health messaging. And with the next slide — some of the work
that we’ve done to really improve our messaging — and messaging, especially, to lay audiences
of some of these complicated One Health dynamics. So, this is one of the infographics that’s
in the report. There’s a lot more details on other pages,
which I’ll get to in a moment. But this is really highlighting some of the
challenges that we face, and I think, have been presented previously, of — when we talk
about One Health, we often put schematics on slides, where there’s about a hundred different
boxes, and there’s arrows going to every single box, showing the interrelated connection. And it’s really hard to understand, and hard
to follow. So, part of what we wanted to do is improve
our communication on it. And this was one way that we saw where we
could try and do that, to highlight the fact that there are a lot of different issues here;
there are a lot of different intersections. As you can see, with all of these little bubbles,
we highlight potential areas where antibiotics are used, or where antibiotic resistance has
been detected, to try and bring awareness for some of those pieces. And on the next slide, you can highlight — you
can see that we sort of dig down in each of these places and sort of zoom in on each of
those parts of the infographic, to provide information and bullets on the impact of human
health, or the impact on animal health, the impact on the environment, and the interrelationship
between all of those pieces. Next slide. The other thing we’d just highlight, as I
conclude — there’s a — two more slides here — is to really note that there are — we
use the report to also — as a sort of a place to jump off, to talk about what else needs
to be done. Obviously, there’s a lot — work that needs
to be done, that the PACCARB has highlighted, and that all the agencies are working on through
the One Health lens. The other things we’d just want to underline
is the global nature of resistance. And it’s really something that I think we’ve
made tremendous domestic improvements in the United States. I think that really, the infrastructure that
we’ve put across the One Health spectrum, and the policy changes, and the programmatic
changes under the first action plan have really advanced the U.S. efforts domestically. I think that we’re really in a challenging
place globally, though. As I like to note to folks, we have a good
understanding of the burden of antibiotic resistance in the U.S. We have a good understanding of the burden
of antibiotic resistance in parts of Europe. We don’t have a very good understanding of
the burden of antibiotic resistance in other parts of the world, and we need better data. We need better understanding of those things,
and we need to understand the interconnectedness of this. Because ultimately, we’re not going to be
able to get ahead of this as a country if we don’t do more on the global front. Next slide. And this is also a page that’s highlighted
in the threats report, where we at CDC sort of highlight all of the different things that
we think needs to be done. This is the shorthand version of it, but we
do have a page in the report that really highlights what we think are the areas of further need
and further exploration. I don’t think any of these will come as a
surprise to the members of the committee, and really, just highlighting that we have
— while we’ve had significant improvements domestically, there are still gaps. There are still things that we need to do. Really, highlighting what I mentioned on the
previous slide, of — there’s a lot that needs to be done globally, not just by the U.S.,
but by, really, every country of the world. And then, on the innovation side, as we talked
about — and spent a lot of time talking about yesterday, there’s a lot that we can do to
leap forward, on both the prevention side, on the detection side, and on the treatment
side — to improve where we are not only as a country, but as a globe, in responding to
this problem. And with that, I thank the committee.>>Martin Blaser: So, Mike, thank you so much. We have a couple of minutes for questions. Does anybody have any questions for Dr. Craig? While you’re waiting, I’ll ask a question,
and that is, so antibiotic resistance comes about because of selective pressure from antibiotics
and other microbes. And you know what? As you point out, one of the big emerging
areas is antibiotic resistance in urinary tract infections. And that’s a whole spectrum, from severe infections
to mild. Fortunately, most of the infections are mild. And the question is, is CDC looking into non-antibiotic
alternatives to reduce that base? Do all people, especially all women with urinary
tract infections, need to get antibiotics?>>Michael Craig: [laughs] Well, so, it’s
a complicated question. On the ESBL side, it’s one where we’re definitely
looking at and trying to understand the causes for those increases. There are certainly things that are related
to the increase of certain types of antibiotics, and we know, for example, things like fluoroquinolones. There can be some selection issues there with
the resistance. But we also know that there’s a lot of other
things going on that we don’t understand really well. And I think, you know, quite frankly, this
is an area of — the burden of UTIs in the United States, as well as around the globe,
is something that has not been really focused on and studied in the depth and need that
it has — that it should be. And it’s an area that we want to do more. I think, on the alternative side, and are
there microbiome approaches — I think we firmly believe that at CDC, and I think the
— some of the conversations yesterday, especially with the presentation by the Gates Foundation
— I think they — those do highlight that there are some non-antibiotic alternatives
that we should be looking at. I think we are very much interested, and a
lot of our SMEs are interested in how we can measure the microbiome. Can you measure disruption of the microbiome? Because if you can measure the disruption
of the microbiome, you can potentially get to endpoints that might be helpful for the
development of a drug, or a therapeutic, or potentially, a decolonizing agent that could
be used as a non-antibiotic alternative.>>Martin Blaser: We have time for one more
question, Dr. Helen Boucher.>>Helen Boucher: Thanks very much. I was just going to add, if you look at the
CARB-X portfolio, there are some very novel ideas about non-drug therapies that may be
relevant. And then, just to shout out to the CDC, I
think one of the biggest things that’s happened — as we saw, the decrease in healthcare-associated
infections — is the addressing of not infections, right — asymptomatic bacteria. So, there’s been a big move in infection prevention
to not diagnose them when they’re not there. And we have a long way to go, but we’ve made
some progress, and I think this is — fits in nicely with what we saw in the threat report.>>Michael Craig: Absolutely.>>Martin Blaser: Thank you so much for all
the progress.>>Michael Craig: Thank you.>>Martin Blaser: Thank you. And now, we’re going to turn to our first
panel, and I’ll introduce Dr. Elaine Larson, from Columbia University, to moderate.>>Elaine Larson: Good morning. For the first panel today, we’ll be taking
a step back and looking at a variety of topics that demonstrate the many different areas
impacted by and contributing to antibiotic resistance. We’ll be looking at the AMR landscape, both
figuratively and literally, and hear some unique perspectives on the issue. Our first speaker is Dr. Paula Cray, Department
Head, Population Health and Pathobiology, North Carolina State University.>>Paula Cray: Thank you very much, Elaine. So, I want to reiterate some of the same themes
that Michael went through, with perhaps, a different perspective [inaudible]. Next slide. I’m here. Okay, so this is a slide I presented a couple
of years ago, at an NAS meeting. And what it does is, it essentially takes
the same themes that Michael was talking about, and it puts it in a One Health perspective. But what I want to emphasize here is that,
if you look at each circle — so, in the top center, you’re looking at the microbial world,
all of the actions and interactions with antibiotics and hormone, hormone-like compounds — so,
everything that can affect the development of resistance. And then, going around the circle, what we
have are all of the areas that actually impact each of us almost daily. So, one of the things that I really want to
emphasize is, is that there isn’t a day that most of us don’t eat vegetables of some sort,
or crops, or grain. We travel. We travel to work, we travel home, we travel
out of the country, in the country. We look at climate. Yesterday, it was warm. Today, it’s cold. We look at birds — and people say, “Well,
I don’t come in contact with birds.” If you touch a handrail, you came in contact
with birds. So, there — people come in contact with birds
daily. And then, we have the migratory bird population. And so, as you look around this continuum,
what we see is, is that in fact, it really isn’t the boxes, like Michael said, where
there’s a uni-direction of where resistance is going or it’s ending up. It is a continuum that affects us every day,
each and every one of us, in every aspect of our lives, from the dogs that we kiss on
a daily basis to the foods that we eat. Next slide, please. So, one of the things that we have embarked
on — and there are two themes in the talk here. Besides the One Health continuum, it’s how
do we look at risk factors for mitigating some of this information? And one of the things that we’ve seen is,
is that every country has antimicrobial resistance reports. They — every hospital has their own reports,
every company has their own reports, and that’s it. The key word is “own.” Everybody has their own. So, what we’ve embarked on is an artificial
intelligence exercise, so that we can collect accurate insights, knowledge of trends in
AMR, and look at particular data sources. And what we’re doing is, there are two approaches. One is from a publication, news media — anything
that’s available online, and the other is to build an actual One Health big database. Next slide, please. Next slide, please. So, what we’ve done first — and I owe a — nearly
all credit goes to Dr. Shiva Keelara, who’s a co-author on this. He — this is his project, and he has done
an amazing job. But what we have is an AMR genie app that
we hope to release at some point in time. And what this is, is it’s actually customized
search engine. And what we’re using is the IBM Watson platform. IBM Watson platform has an amazing ability
to read papers — tens of thousands of papers in minutes, and sort the information. So, once you go through the training, we can
develop customized algorithms so that — and in fact, we’ll use the data sources that would
be most accurate. And what do I mean by that? A public search engine — if you go to Google,
Bing, Yahoo, and you put in “salmonella Heidelberg” — next slide, please — you’ll see that — and
the genie is actually running now. So, what we did was, we asked it, “What are
the common sources of resistance in cattle,” okay? And of course, it comes back with the first
highlight that says, “Cattle are the first.” Which is good, because that’s proof of concept. It got all the cattle. And it goes through, and it gives you article
after article. It’ll give you the citation; it’ll send you
where you need to go. Versus, on the right-hand side, you’ll see
that Google search — if you say, “What are the common sources of resistance,” the first
thing it comes up with — resistance to change. So, it does a common search using just the
keyword “resistance.” So, by using this, we hope to be able to put
it into peoples’ hands, so that they can become educated quickly. And you could, perhaps, tailor this to use
in the field in some way, and doctors may be able to use this for, like, latest treatment
sources and various other things — applications. Next slide, please. So, in — with respect to the big database,
though, is that we know that every country has reports. We have NARMS in Denmark, NORM-VET in Norway,
and so on, and so forth — all of the hospital information. And one of the most amazing things about using
the Watson platform — Watson runs our weather system right now. And it updates the weather every three minutes
globally, okay? So, that’s why you can go on any state and
see the change in a matter of minutes. So, we’re taking this platform. This platform is going out and it’s taking
all of the static information from these reports and putting it into a single database. And you’ll notice that there’s a discontinuous
line there, and we know that some of the sources for resistance and some of the information
is, in fact, going to be discontinuous. But what we can do is, is we can overlay any
source of data on top of the resistance data. For instance, migratory bird patterns — when
do they run? What are there — the bacteria associated
with them? What’s the temporal pattern? We can run weather; we can run humidity, temperature,
wind, whatever parameters that are captured — travel information, other infectious diseases. Typically, bacterial infections are secondary
disease, and my goal is to plot viral disease, and see what comes up with the secondary bacterial
diseases, and then, look at resistance associated with them. And the arrows depict how you can look at
this. You can look at it across all of the databases
and ask big, broad questions, or you can look at it on a single day, okay? So, we — this is taking an inordinate amount
of time, but we think that when it’s finished, it will assimilate all of the complimentary
and supportive data and facilitate a rapid diagnostic development in generation of new
knowledge. Next slide, please. So, and this is a common theme that I want
to reiterate, is we have, as researchers, five unique opportunities to study AMR and
how it impacts us in the environment. Every time there’s a hurricane, a flood, a
drought, a fire, or a volcanic eruption, in each case, the environment in proximity to
the event will reset, okay? So, you’ll get ash cover everything and set
up a whole new microbiota or microbiome in the soil, essentially. Same thing with water. And if you look up in the top, right-hand
corner, you’ll see that flooding — I — it’s interesting you talk about UTIs. One of the big events when we have floods,
particularly like after a hurricane, is all of the sewage mixes from all of the houses. All of the women who have UTIs, all of the
whatever is in the home sewage is now mixed and mingled with everything from animal production
to wildlife and on. So, what we did, then, is — next slide, please
— is another — I keep Dr. Keelara busy sometimes with, “Hey, let’s — this is a good idea.” But you’ll notice, on the left-hand side — this
is what happened after Hurricane Florence. And the red — if you’ll look from the inlet
— the black is the water. And if you go up to the red, you’ll notice
the branches of deep red. And those are areas that actually flooded
along the Neuse River, okay? So, that’s a 30-mile stretch. What we did, you’ll see on the right-hand
slide. We took six different areas. We took a residential area, a poultry area,
a — close to a swine farm, a hospital, and a recreational area at the outlet, and we
looked at all of these sites, took soil and water from these sites. And we’re just analyzing all of the data that
we have now, as we have hundreds and hundreds of isolates. And we can see that — next slide, please
— that interestingly, are salmonella — and we have salmonella, essentially, from every
site except site five, which was around the human hospital area. We only had one isolate from soil. None from the water, which was quite surprising. But you’ll notice that the — a lot of these
isolates are pansusceptible, or only resistant to streptomycin, and I found that particularly
striking. Whereas, on the bottom, you’ll see that — those
are ESBL E. coli isolates, of which, we were able to isolate 67 along the continuum throughout
the year. And you’ll notice the inordinate amount of
resistance — multi-drug resistance — from these E. coli. And essentially, these E. coli were isolated
from each of the areas along the continuum, and from both soil and water. And what this — you know, what this says
is, is that sometimes, when we make suppositions, they’re not always accurate. [laughs] So, people will say, “Well, you’re
going to get all of your resistance from your salmonella.” And it doesn’t look like that, in fact, is
true, even though we know we have 30 miles of salmonella saturating the water and the
soil along the Neuse River. Whereas, if we look at a specific population
of E. coli — and granted, we isolated and selected for ESBL E. coli — but we see the
same type of resistance from generic, or E. coli, too. So, the populations are quite different, and
we hope to do much more analysis on this, so that we can have a final story for you,
hopefully by the end of the year. Next slide, please. So, that’s it. I just want to remind you that we are in this
continuum, and we are what we do every day, starting from the moment you wake up.>>Elaine Larson: Thank you, Dr. Cray. Our second speaker is Anna — Dr. Anna Dhody,
Director of the Mutter Research Institute.>>Anna Dhody: Good morning. I would like to take this time to thank the
committee for their invitation to speak at this meeting. I hope to offer a unique perspective in the
fight against AMR, and inform the larger scientific community of the importance of museums and
their historical biorepositories as valuable and largely untapped resources. The Mutter Museum is part of the College of
Physicians of Philadelphia, one of the oldest professional medical societies in the United
States. Our mission is to advance the cause of health
while upholding the ideals and heritage of medicine. This was true in 1787, when we were founded,
and it is still true today. However, one of the ways we are fulfilling
our mission is unique. The Mutter Museum opened in 1863, and contains
thousands of historical medical specimens. The museum strives to fulfil the mission of
the college by educating our visitors about health and the human body. For the first century or so, the museum was
almost exclusively a resource for the medical community. However, in the past 40 years, the museum
has welcomed and encouraged all visitors, and today, receives over 180,000 visitors
annually, the vast majority of whom have no medical background or training. In 2007, we embarked on a collaborative project
with the Ancient DNA Centre in McMaster University. This resulted in the first successful extraction
of ancient pathogen DNA from a fluid, preserved specimen. The specimen was a section of intestine from
an individual who died of cholera in Philadelphia in 1849. Using a piece of intestine smaller than a
nail clipping, we were able to reconstruct the entire genome of this second pandemic
cholera strain, and identify key virulence factors that played a role in the outbreak. The response to these publications led to
the creation of the Mutter Research Institute. The goal of the Institute is to further our
mission by collaborating with scientific institutions and researchers to engender knowledge about
the body and health, and to increase our understanding of medicine and the evolution of pathogens
within a cultural and historic context. The Institute currently has over 10 active
projects in various phases. The one I am here to talk to you about today
involves a rescued cemetery population, dental calculus, and the oral microbiome. Next slide, please. Next slide. Oh, back one slide [laughs]. In late 2016, myself and a colleague from
Rutgers-Camden visited a construction site on Arch Street, in Philadelphia, and we were
given a box of human bones — the remnants of an 18th century cemetery that had been
relocated in the 1850s. We knew the cemetery where the inhabitants
were supposedly relocated and offered to assist with the few still left at the original site. I think we all know that saying, “No good
deed?” In early 2017, we received a frantic call
to go back to the construction site, where coffins were piled up three and four-deep. It turns out the inhabitants were not relocated,
as we had previously thought. We knew we had a responsibility to remove
and relocate these individuals to the best of our ability. From March until September of 2017, a total
of 491 sets of human remains were excavated by our team, and then, a CRM firm. We worked with the Orphans’ Court of Philadelphia
and the descendant community of the church where they were buried, and received permission
to study, sample, and analyze the remains before they are reburied in 2023. As a museum curator, I am interested in both
conducting scientific analysis with museum specimens, but also, preserving biological
materials to the greatest extent possible. Thus, I’m not overly keen on destructive analysis,
unless it has tremendous potential to uncover the past in a meaningful fashion. I was interested in seeing what we could learn
from this population without having to sample bone, tooth, or other tissue. I had read about dental calculus as a potential
source of not just the DNA of the individual, but also, of their microbiome. The Arch Street population had a significant
number of individuals with dental calculus, and as the sampling does not damage the teeth,
I asked my collaborators at McMaster for a little favor. Next slide. Would they analyze just a few samples for
me? They agreed. Sampling of the material was completed at
the Mutter, in our conservation lab. Calculus was removed with the sterile dental
tools, and small pieces were collected for analysis. The calculus and a couple of opportunistic
soft tissue samples were transported to McMaster for further processing and analysis. Mineral and protein components of the calculus
were broken down, to allow for DNA purification. As preliminary pilot project, the DNA libraries
were shotgun sequenced to ensure there was no bias in the data generation, and so we
could get a broad overview picture of the composition of the calculus samples. We compared the DNA sequences to a database
of reference bacteria and viruses, to try and identify the organisms. We also mapped our DNA reads against the human
nuclear and mitochondrial reference genomes. However, in both cases, the human fraction
was too low to assign a mitochondrial haplogroup or determine sex. However, with deeper sequencing or targeted
enrichment for the human components, we believe there would be no difficulty in recovering
these molecules from the calculus samples. Next slide, please. In this figure, you can see some of our initial
results. The size of the bubbles correlates with the
number of molecules identified to selected taxa in each sample. There is a clear difference between the blanks
and the samples of calculus and soft tissue. We can see abundant oral bacteria in the calculus,
though not always the same taxa, or in the same proportions. As expected, the soft tissue samples have
next to none of the oral taxa. This is a good sign, and indicates the molecules
recovered from each sample are specific to the source material, not the source individual
or contaminating environment. The blanks, again, look different — no oral
taxa — and none of the soil taxa most common in the soft tissue. We see that the blanks are dominated by Bradyrhizobium,
consistent with the profile of lab reagents, meaning no contamination DNA from an external
source. Just a note about the contamination: we’re
using the term to refer to introduction of DNA from the soil into the calculus and soft
tissue as a natural process, rather than contamination introduced during handling or storage of the
samples. This process shows we are capable of recovering
the oral microbiome from these calculus samples. We found additional taxa in the oral microbiome
that indicates the presence of obligate pathogens. However, the presence of a pathogen’s DNA
does not mean it was the cause of death. One of the complexities of oral microbiome
studies is that several taxa identified can be both part of the normal flora, or it can
be opportunistic pathogens. We are excited about continuing this research
and have just started following up on this information. Next slide, please. So, how is all this significant in the study
of, and potential help with, AMR? We have a limited timeline of known AMR, as
well as what microbes are carrying AMR genes. We feel that looking back at the populations
pre-1950s is going to be crucial to see what genes were already being passed around in
humans prior to the selective pressures of antibiotics in the late 1940s and early ’50s. Our collaborators in McMaster have designed
and tested a bait set for all currently-known human-associated resistance genes. The individuals in the Arch Street population
date from approximately 1707 to 1858. We know from the historical records that there
were significant epidemics of infectious disease during this period, such as cholera, smallpox,
and the yellow fever epidemic that killed approximately 5,000 people in Philadelphia
in 1793, or about 10 percent of the population of 50,000, at the time. The age of many of these individuals indicate
they lived through these outbreaks. Perhaps there is something in their microbiome
that can inform us why. Next slide, please. The additional significance of these results
are many. Our initial analysis shows the DNA preservation
in calculus is excellent. The metagenomic profile is very distinct from
both the soft tissue and the extraction blanks. There is less contamination in the calculus,
and the contamination we see is from soil, not from the handling or storage of the samples. We believe that there is significant potential
for further analysis of this population from multiple sources, such as oral microbiome,
human host, food sources, soil samples, and the potential pathogens. We know we’re getting oral taxa, and that
there are many more samples to study. So, population-level analyses are a possibility
for the future, and the statistical comparisons from population-level analyses could tell
us much more than just our small-scale pilot study is able to. There is a hole in the literature, in terms
of looking at dental calculus from the time and place that these individuals originate. All of this data can inform the medical community
about human health in the pre-antibiotic world. At the very least, we have proved we can recover
high-quality DNA from dental calculus from this site. Next slide, please. My colleagues and I are eager to continue
this work. Moving forward, the endogenous DNA will allow
us to transition into further specific study of individual taxa and species that are present
in the samples. We plan to look at the soil samples from the
site as another point to show that the calculus has its own unique, authentic signal. We need to develop strong research questions
that will make use of this data, and of course, explore funding opportunities that will allow
us to conduct further analysis on additional samples. This population contains a potential biorepository,
the breadth and scope of which we have not yet begun to fathom. The knowledge we can gain from these people
may help inform us how to manage AMR. Who knows? The answers may be there. It is up to us to ask the questions. Next slide. These are just our references. Next slide. And I would just like to thank my fellow Arch
Street colleagues, and of course, our long-term collaborators at McMaster. And while I may think long and hard about
accepting one box of bones ever again, I am grateful to be a part of this amazing project. Thank you.>>Elaine Larson: Thank you, Dr. Dhody. Fascinating. So, our next speaker is Dr. David Khan. He’s the — on the board of directors of the
American Academy of Allergy, Asthma, and Immunology.>>David Khan: I want to thank the council
for this opportunity to speak on this listening session for antimicrobial resistance. And many of you may be wondering why an allergist
immunologist has any interest in this topic, but hopefully, during my presentation, that
will become quite clear. I would like to take this opportunity to introduce
you to the organization I’m speaking on behalf of, the American Academy of Allergy, Asthma,
and Immunology. We have 7,000 members in the U.S. and Canada
and 72 other countries, and our mission is dedicated to the advancement the knowledge
of allergy, asthma, and immunology for optimal patient care. And it’s — this part, the optimal patient
care, that leads me to talk about the importance of penicillin allergy de-labeling. So, all of you are aware that penicillin allergy
is a common allergy that’s reported amongst patients. On the top, right-hand corner of the slide,
this is looking at a database from Boston, over 1.5 million patients. Looking at rates of antibiotic allergy — and
the blue line, at the top, is penicillin, which clearly dwarfs the other antibiotics,
and is the most common allergy of any kind — antibiotic or otherwise. However, most patients who carry this label
of penicillin allergy, when evaluated, are really not allergic. The graphic here is showing longitudinal data
looking at penicillin skin testing as an outcome for a positive test. And this is from Kaiser’s data. And you can see, in the mid-1990s, maybe about
15 percent of patients who had histories of penicillin allergy would have a positive skin
test. As we fast-forward to about 2013, that number
is less than 1 percent. Recent data from the Mayo Clinic, looking
at over 30,000 patients that they’ve tested with histories of penicillin allergy — the
rate of positive skin tests is less than 1 percent. So, clearly, the bulk of these patients with
allergy label are really not allergic. Next slide. So, why do we care? Why does it matter if someone has this allergy
label? You know, there’s plenty of other antibiotics
one can use. Well, it’s because this does come at a cost,
both for the individual and the society. Most penicillin allergy labels occur during
childhood. And we now know that most of this is because
they’ve been inappropriately labeled as being allergic to the drug, when their rash is probably
related to the viral exanthem they had. From a personal perspective, patients who
get other, alternative antibiotics may experience more toxicity. They’ll get more broad-spectrum antibiotics,
and they can have worse outcomes, because penicillin might be the drug of choice for
that particular issue. From a public health perspective, this certainly
leads to antibiotic resistance. We see higher rates of C. difficile, increased
hospital stays, and a number of other issues. And again, when we evaluate patients, the
majority of them are really not allergic. Next slide. Now, when using other, alternative antibiotics,
as it turns out, that may be — not be the best thing. Penicillin and other beta-lactams are oftentimes
the drug of choice for a number of different infections, and one can see better outcomes
when using these antibiotics, as opposed to the alternatives. Next slide. The morbidity associated with this label of
penicillin allergy is something relatively new. It’s something that we didn’t really appreciate
until relatively recently. Here are two large studies involving 50,000
cases of patients who have labels of penicillin allergy, one from the U.S., from Eric Macy
at Kaiser, and another, looking at a U.K. database published by Kim Blumenthal. And both of them showed that these patients
labelled with penicillin allergy were at higher risk for development of MRSA, C. difficile
— very important infections. More recently, data from Kim Blumenthal looking,
again, at a large healthcare database of general practitioners — about 2 million patients
— found that the mortality was actually higher — 14 percent higher mortality from carrying
this penicillin allergy label. Next slide. So, what can we do about it? So, when you evaluate patients for penicillin
allergy and you remove that label, we call that “de-labeling.” And does that have an impact? As it turns out, it does. So, again, another study from Eric Macy, showing
that de-labeling patients from penicillin allergy reduced their healthcare utilization
— less hospitalization, less E.R. visits — and that translated to costs savings of
around $1,900 per patient. In the lower figure, what we’re showing is
data from our own institution, at Parkland Hospital, where we have developed a system
where we train pharmacists how to do penicillin allergy testing in the hospital. And so, they go around and identify these
patients, and are able to de-label them in that way. What is shown here is reduction of broad-spectrum
antibiotics like Vancomycin, fluroquinolones, and then, increase, obviously, in penicillins. Next slide. Now, the CDC has issued this fact bulletin,
in 2016, indicating some of these facts that most patients who think they’re allergic to
penicillin really aren’t, and did advocate for evaluation for penicillin allergy. In 2017, our organization, the Academy, issued
this position statement, basically saying that penicillin allergy testing should be
performed routinely in patients with self-reported penicillin allergy. This is a real sea change for us, as allergists,
because in the past, we would only evaluate penicillin allergy when there was a real need
for penicillin, not for other reasons. So now, we’re trying to do this as routine
care. So, when I see patients who come in for their
allergic rhinitis, or their asthma, or other reasons, and I find out that oh, yeah, here’s
a penicillin allergy label on their chart, we try and actually address that, even though
that’s not why they’re coming in. Next slide. So, how can one remove the label of “penicillin
allergy”? Well, there’s a lot of different ways, and
we don’t really have time to go into this. But I do want to kind of point out the different
settings that this can occur in. So, this can occur, clearly, in the outpatient
setting, as well as the inpatient setting, ranging anywhere from the emergency room,
even to the intensive care unit. Various programs have shown that you can do
penicillin allergy testing safely in a number of different situations. We generally advocate to do this proactively,
to remove that label, so that there’s no delay in appropriate therapy, especially for hospitalized
patients. Next slide. So, one of the things — or, several of the
things that we’d like to have kind of a call for action here would be outreach — not only
to providers, but really, to the public at large — about the importance of de-labeling
people with penicillin allergy. There are certain target populations who may
benefit even more: patients with cancer, transplantation, diabetes — those that need more antibiotics. Even pregnant women would benefit from removing
that label of penicillin allergy. We also want to encourage the FDA to support
our complete penicillin skin test reagent. So, right now, the testing that’s commercially
available does not have all the reagents, and this raises some concerns about safety
of testing. And that would be great, if we could have
all the things that would be necessarily to de-label patients in a little more safe manner. Increasing reimbursement for testing, I think,
will encourage providers to actually do this work and help de-label all these folks. And then, I think, as the title suggested,
adoption of a penicillin allergy testing program should be a component of antimicrobial stewardship. Here’s our email, to contact us for any information
that you may have in this regard, and we’d certainly be happy to do — to address any
questions from any of the stakeholders here. Next slide, please And I certainly want you to all put in your
calendars this date, which is September 28th, National Penicillin Allergy Day. So, you can celebrate with all of us about
this day. And why September 28th? Well, that’s the day that Alexander Fleming
discovered penicillin. So, thank you so much.>>Elaine Larson: Thank you very much, Dr.
Khan. Our last speaker in this panel is Dr. Randy
Singer, Professor of Epidemiology, University of Minnesota, Department of Veterinary and
Biological Sciences.>>Randy Singer: Great. Good morning, and thank you all for the invitation
to speak to the council today. It’s really nice to be back and see you all. Today, I’m going to talk about the progress
that’s being made on investigating AMR in the natural environment. I’m going to trace a little bit of my history
in this area, as well — a couple projects that we are currently performing in this area. And hopefully, in this brief topic, we’ll
be able to give you a quick update on where we are with the natural environment. Next slide. So, I’m going to begin with this graphic that
came out in a paper of 2014. I think it does a really nice job of framing
AMR, or antimicrobial use, in a One Health context. You know, I — maybe I’ll consider switching
to the CDC graphic that we saw earlier, because that, also, I think, did a nice job. But what you see in this graphic is where
antimicrobials are being used in both the human and animal aspects. You — the red dots might represent the release
of antimicrobial metabolites into the environment. And I think the importance of this image really
is about how the antimicrobial ends up being concentrated in different compartments of
the environment — of the natural environment — and that they’re all truly in an interconnected
space. The thing about the graphic is, it really
focuses, though, on the use of the antimicrobials in human and animal populations. Crops are represented, so we need to remember
that we do have antimicrobial use in crops. What’s not presented would be things like,
pharmaceutical manufacturing, the use of antimicrobials in ethanol plants. There are many point sources that would also
be contributors to the release of antimicrobial metabolites in the environment. What I want to focus on, though — and I think
it’ll be a theme that I’ll use in this presentation — is a quote from the paper, where they state
that the life cycle of pharmaceutically-used antibiotics does not simply end when a patient
swallows a pill or when livestock are treated. They really go into some detail about the
chemistry of the antimicrobials that are being released, those physicochemical properties
that allow the metabolites to bind into organic substrates in different ways. And that’s what makes predicting the concentrations
of antimicrobial metabolites, and of resistant bacteria and their genes in the environment,
a challenge for us. Next slide. So, I’m going to jump back. We’ll walk back in time a little bit to a
paper I put out in 2006, where we tried to set the stage for, how would you more accurately
relate antimicrobial use and resistance, especially within the context of that natural environment. What kind of analyses do we need to do to
account for the complex relationship of use and resistance and the many sources of antimicrobial
resistance into the natural environment? I’ll briefly just look at those four graphics
on the right. I obviously am not an artist. This is actually in the manuscript itself,
but what you can see in these graphics. I’ll start with the top left would be imagine
you have two different populations: in this case animals, in the environment. Maybe the populations receive a different
antibiotic for the same condition, you might expect then to see resistance correlating
to the antibiotic that’s being used in each respective population, but it’s actually not
that simple of course. On the top right what we see is that there
could be many features in the environment that actually affect which bacteria and which
genes are going to be more likely to persist in that environment, for instance, maybe it’s
a distribution of copper that might be in the environment that helps coselect for microbial
resistance. A geospatial method, which we were trying
to explore in this paper, might help you account for that distribution that doesn’t fit exactly
with which antimicrobials are being used. The lower left we can introduce point sources
of AMR into the environment; it could be hospitals, wastewater treatment plants. On the right — bottom right we have the spread
of AMR as well, it could be airborne dissemination. You tend to be more similar to your neighbors
so you should probably take clustering into account when you’re looking at the spatial
relationships. But the point, again, of this paper was too
how do we use principles of landscape ecology, geospatial analysis, to better characterize
the AMR and antimicrobial residue findings in the natural environment. Next slide. So what — a really thorough and one of the
first studies that tried to do this, I think, was in 2012. This paper was put out by a Spanish team that
tried to estimate where would we expect to see antimicrobial metabolites loading in the
soil in Europe. This was a — is a really interesting paper
and they tried to characterize it using a risk assessment approach, so really that is
four steps; they talked about a release assessment, an exposure assessment, a consequence assessment,
and then a risk estimation. What you’re seeing on those figures than would
be first, on the release assessment, they used livestock density and actually they only
used beef or cattle and Swine livestock density as a indicator of how much antimicrobial might
be used. So they had some antimicrobial use data by
species and with that they’re trying to use a proportionality assumption to say how much
antimicrobial might I expect to be used across this landscape. The exposure assessment then would say how
much of that is going to be released into the environment and based on the individual
properties of each of those antimicrobials, how likely is it to persist in the soil, so
the physical chemical properties of the environment. The consequence assessment would be how is
the soil that each of these locations being used; is it is it a crop production facility? So maybe then your risk might be greater. And then finally they call it the risk estimation,
which in my opinion is not really a risk, it’s more soil vulnerability. So where would you expect to see loading of
antimicrobial compounds across this broad landscape. It was one of the first papers that really
brought these geospatial tools into this context. Next slide. So what we are doing currently in Minnesota,
and this is part of our Minnesota One Health Antimicrobial Stewardship Collaborative. In 2019 Dr. Amanda Beaudoin presented to PACCARB
on this topic. We have a diverse — very diverse group of
individuals in this state. We have three different — several working
groups. I tend to be on this one mainly called “The
Antibiotic Footprint Group” and in our team we formed a project around this overarching
hypothesis that we could build a spatial model for Minnesota that would predict where we
would find a AMR in the natural environment as a way to guide intervention. We had several objectives, really though,
about building a geospatial model and one in which we would validate it by sampling
in the environment as well. Next slide. And so the way we’ve approached this, and
this is still a pretty early on, is what you see on the left where it says antimicrobial
loading. We work with densities, so it’s hospital densities,
wastewater treatment, the various agricultural facilities around the state. Again, we have antimicrobial estimates of
use both in the human and animal populations that we layer on top of that to try to get
an idea of where we would expect to see antimicrobials being loaded into the environment. We then add in in the middle different kinds
of geospatial attributes like land cover, surface water, because again, those properties
help us predict where we would find resistant bacteria, resistance genes, and those metabolites. So then we combine that with the field sampling
that we’re doing on the right, you can see a couple of the antimicrobials that we look
for in the water supplies around the — around the state. We’ve modeled here for you enrofloxacin and
sulfa drugs so we can then try to predict where would these concentrations be the highest
based on the observations and the rest of that geospatial model. We’ll build this model for both the resistance
genes and for antimicrobial metabolites, and then we go back out to the field at sample
locations that we — weren’t previously visited and validate the model, update the model,
and hopefully have a model that allows us to identify hot spots and appropriate risk
mitigation activities. Next slide. So there’s another working — within our footprint
working group, there’s another team and their names will be listed on the acknowledgement
slide, that tried to address antibiotic footprint in a different way. This comes more from the human side. The — there was a paper published recently
on an antibiotic footprint as a communication tool. Dr. Laxminarayan here — who was here in this
— on PACCARB was one of the co-authors on this paper. The paper that was really based on consumption
of antibiotics and, you know, if we were really did think about a carbon footprint, it’s more
than just the production of carbon, it’s about total emissions which would include things
like the sources, but also sinks and long-term persistence that you might observe of carbon
in the environment. So what we tried to do on this antibiotic
footprint piece is expand — it’s not just about consumption, but it’s also about metabolism,
it’s about excretion, and then it’s about environmental persistence. So on the right you kind of see some of the
ideas that were taking into account as this group puts together an antibiotic footprint. Again, this is mostly based on human pharmaceuticals. Next slide. So within the last 10 years, I mean, there’s
been an explosion now in the number of papers that you’ll see that address this topic of
AMR and of microbial metabolites in the environment from the top-left where Dr. Amy Pruden who
presented here at PACCARB previously back in 2012, was looking at resistance genes in
the environment. Interestingly, you had differences based on
the gene of their ability to correlate to environmental features, highlighting this
idea again that you can’t just put one model that covers everything in the environment,
it’s a very specific thing for the drug, for the bacterium, for the gene, and that specific
environment. In the middle was another exercise published
in 2019 that looked at mapping of different aspects of AMR and of antimicrobial metabolites
in the environment. It was not based on sampling though, so this
was more of an exploratory GIS project. On the bottom right was an interesting study,
very thorough, about the dissemination of antimicrobial metabolites into the environment
in the River Thames, and what’s interesting about it is that this modeling approach took
into account the properties — the physicochemical properties of the environment, and human prescriptions
of different drugs, fluoroquinolones and macrolides, to predict how high of the concentration,
and how far you would see these elevated concentrations in these waterways. Next slide. So to conclude, I mean, what makes these projects
really difficult and trying to have accurate predictions of where you’ll find hotspots
is that it really varies by the gene, by the drug that you’re actually investigating, and
is relationship to that specific environment, and so we’re really work hard to understand
it at a smaller scale. We’re working just in Minnesota and even that
is a very large landscape. But the bottom bullet, I think, is really
where we still struggle. How do we take the environmental findings
and relate them back to health impacts? You know, this paper published by Ashbolt
et al. in 2013 set a framework for how we might do these kinds of risk assessments that
incorporate the environmental AMR component, but I would say that this area is still in
its infancy, and that really is, I think, where we struggle the most. How to relate it back to the health impacts. Next slide. And so I would just like to acknowledge the
collaborators on the right, the people that are working on that other footprint model
related to human prescriptions in Minnesota, and with that, I guess, we take questions. Thank you.>>Elaine Larson: Yes, thank you very much. Thanks to all the panelists and while you
get your signs up for asking questions, we have about a half an hour for questions, I’ll
start with Dr. Khan. A colleague and I just surveyed a number of
nurses in several hospitals about how they — because they’re the ones who write down
there’s an allergy, a penicillin allergy, and how they do it. And of course, all they do is say, “Are you
allergic to penicillin?” Yes, no, and then they write it down. It seems to us the most efficient way to — would
be to really screen out some who are clearly not allergies, would be to have a screening
tool that the nurses at admission to the hospital could ask patients couple of questions, and
if those questions are clearly like, “Oh, one time I — my, you know, my skin turned
red or something. I don’t know.” That seems like a better way or most efficient
way, even before asking the pharmacist to go make rounds, so we are now working on some
ideas about that, but it would be great to know do you think it would be safe and appropriate
to have a screening tool for the nurses to just use when patients come into the hospital?>>David Khan: Yeah, thank you. That’s a great comment and idea. Certainly a number of these patients were
labeled with penicillin allergy do not have histories that are even remotely suggestive
of an allergy, and I think — and we’re in the process of, kind of, doing that at our
own institution as well to, you know, kind of, walk through some of the things that you
could clearly delabel without any concerns. For example, I had a headache when I took
penicillin, I had diarrhea when I took penicillin, my mom is allergic to penicillin, all of these
things we identify as a no-risk, and I think that’s fine for anybody, any healthcare personnel
to be able to, kind of, walk through the algorithm and say okay, we can remove that label. Part of the struggle with removing the label
is that it tends to come back. So a number of different studies have shown
that anywhere from 25 to 50 percent of patients who after they go through allergy testing,
that label pops back up, and there’s a number of reasons for that but a common one is that
a lot of these people have had this allergy since they were kids, and for the last 50
years they’ve been regurgitating I’m allergic to penicillin, and it comes for the brainstem
level and they can’t help themselves. And so that’s part of it and then, you know,
sometimes with the amalgamation of EMRs that, you know, one says you aren’t and another
one is, so there is a little bit of an education that also is involved, but I think you know
there’s only, you know, roughly 6,000 board-certified allergist in the US. There’s 35 million patients who are labeled
as penicillin allergy. There’s no way as our specialty can handle
this alone, I think that’s really low-hanging fruit that from a system-wide perspective,
we can get, you know, 10, 20 percent of these right off the bat. I agree with you completely.>>Elaine Larson: Thank you. We’ll start over here. Ramanan?>>Ramanan Laxminarayan: Thanks, Elaine. So question for Dr. Singer. So that’s fascinating work and I agree with
you this is still in its infancy, and it will grow a lot as we understand how resistance
genes move through. Have you had much of a chance to see what
the folks in the sewage [unintelligible] project are doing? Is there a possibility that we then take this
work to the next set of applications, which is if we can say, for instance, as others
have done, look at hospital resistance and use it as a proxy or a measure of actual clinical
resistance in a hospital. One example or second to, perhaps, inform
you know, how we might want to treat wastewater or what the point of interventions might be. So all the stuff is at the point of understanding,
but have you thought about how we can use this to then get to the point of intervention,
to know which interventions are worthwhile and which ones are probably not worth the
expense?>>Randall Singer: Yeah, thanks Ramanan. I think that actually gets right to the point. It’s not just about the human health impacts
but also about how do we intervene, you know? After I submitted my slide set, at a systematic
review was just published about a week ago looking at interventions for environmental
AMR. Most of the papers that were retrieved in
the systematic review were related to interventions around wastewater. Intervening in other aspects of the natural
environment compartments are much less studied, with interventions being much less clear how
you would approach it, but this paper did a nice job, I think, of laying out some of
those wastewater interventions that seemed to work for decreasing the concentrations
of the genes in the effluent that might then enter the waterways. What I don’t see as much of are the interventions
that would reduce the amount of the antimicrobial metabolites themselves. Some these interventions seem to work well
against the bacteria in the genes, not so well against the pharmaceuticals. So that is where we’re going, I mean, I didn’t
— I was more focused on the hotspot piece, less on intervention, but ultimately that
is the point, and I would say that there are huge data gaps on non-wastewater aspects of
environmental AMR that — which strategies would you employ, and is it even worthwhile
because we don’t have a way to really predict what the health impact is, so what would the
reduction in a health — negative health effects be by intervening? Which I think anyone who puts an intervention
into place would want to know what that return on investment is. But related to your question, though, of could
you predict hospital, like, clinical resistance based and what you might find in that wastewater,
that has been a challenge for us. I’m not a wastewater treatment plant expert
but we do have those on our team, but that hospital effluent seems to often get mixed
with effluents in the community as well, and so how do you partition where would you sample
to make sure that you’re really looking at the hospital component only, but I think that
actually might hold some potential if you can find the right sampling point to see an
aggregate, what types of resistance genes are being, I guess, are circulating within
the hospital.>>Ramanan Laxminarayan: I could just quickly
ask, who’s funding this kind of work?>>Randall Singer: So, yeah, our project specifically
and I knew I was a little short on time, so I skipped some of it. It comes — ours comes from the legislative
commission, it’s a fund within the state of Minnesota called LCCMR, which actually is
how Minnesota is using its lottery money, it puts it towards environmental causes, and
so our funding comes from that to address this environmental aspect of AMR. We collaborate heavily with, like, Minnesota
pollution control and other groups. Nationally or internationally, you know, I
think it’s always a challenge to find funding for studying the environmental pieces because
it’s not really necessarily health-related, it’s not necessarily animal related and, you
know, so I think it is tricky, but we’ve been lucky in the state of Minnesota to have that
opportunity. Thanks.>>Aileen Marty: Thank you. Those were outstanding presentations. And my question is for Anna Dhody. So I have a soft spot for anthropology, so
I greatly appreciate your work but that said, we know that AMR genes have been identified
in frozen sediment cores and resistance predates our use of antibiotics somewhere between 2
billion and 30,000 years. So I have to ask have you searched specifically
for AMR genes in your samples, and if so, what similarities and differences have you
found relative to the current problems with antibiotic resistance, and specifically what
have you learned regarding the epigenetics of these genes and microbes that allow the
genes to be expressed and thus thwart our attempts to rescue patients infected with
resistant bacteria?>>Anna Dhody: Thank you for this question. I should mention when I said it was a very
small pilot project that we started with, out of the 491 individuals, we just did 11
of these, and we just did it a couple months ago. It was a pilot project literally just to see
what the calculus could offer us so these are, as I mentioned in the last thing, we
have a lot of questions that we have to ask and address and this is — and those are definitely
some of them. So our next step is to talk to the members
of the community such as yourself, as well as the anthropological community, and come
up with these questions, hopefully find some funding, and then try to address them.>>Elaine Larson: All right. Dr. Blaser?>>Martin Blaser: Thank you. Thank you to our speakers for a diverse and
interesting program. My question is about penicillin allergy and
directed to Dr. Khan, and maybe to CDC as well. So in the — what everybody’s worried about
with penicillin allergy is fatal anaphylaxis and the question is what are the numbers? In the 1950s there were large eradication
campaigns in Africa run by WHO and in China to eliminate syphilis, in which millions of
people were eradicated, and from those studies it looked like the fatal anaphylaxis trade
was about one in a million administrations. So it’s, kind of, like, the reverse of winning
the lottery. So the question is are there more recent data? Are there data in the United States about
how often fatal anaphylaxis occurs and what’s the rate per administration, and does the
route of administration make a difference?>>David Khan: Yeah, thank you for that question. There is more recent data about fatal anaphylaxis
just in terms of drug-induced — drug-induced fatalities from anaphylaxis. Penicillin is still the highest of the drugs
that cause a fatal anaphylaxis, but you’re right in terms of frequency, it’s quite modest. There’s an estimation that from the U.K. about
oral amoxicillin and risk for fatal anaphylaxis, and it was estimated in the range of 1 in
100 million doses, so very, very low. Data from Kaiser suggests that for parenteral
causes of — and this was more for anaphylaxis, not fatalities, but parenteral rates were
much higher than for oral administration. In general, when — with anaphylaxis of any
cause, and I would say drug anaphylaxis is in there, the fatality rate is less than .1
percent of someone who has anaphylaxis will then have a fatality. So you’re absolutely right, we make a big
deal out of, you know, worrying about the fatalities and it’s, you know, the kind of
mindset of you do a lot of things to protect those rare individuals against harm and it’s
a balancing act.>>Martin Blaser: Yeah, and that’s really
my point is that’s where we need data, because it might be by substituting non-penicillins
we’re doing more harm than good. To prevent an infinitesimally rare event,
we’re treating with agents that have a variety of other side effects and untoward complications,
so the perfect is the enemy of the good.>>David Khan: Yeah, absolutely.>>Elaine Larson: Dr. King?>>Lonnie King: So thank all of you, very
much. Dr. Singer, this is for you. So I agree that you know the real gap is exposure
and risk and we have a lot to do to, kind of, understand what — how to fill that gap
but what I want — you’re thinking about is the possibility of using your technology or
methodology and moving it to low and middle-income countries where some people believe that maybe
sewage the best source for surveillance. So can we start using some of this methodology
into some really, you know, rapid, effective, cheaper surveillance worldwide to get a better
idea about AMR. Ramanan, you might have some ideas?>>Randall Singer: Yeah, I mean, I think there
are any members of this council who could also weigh in on this for sure. Thank you for the question, Dr. King. What we see routinely in the literature is
one of the most significant inputs into the natural environment is coming through wastewater
for sure. As well, in these geospatial modeling, many
of these data are publicly available that even in some of the low and middle-income
countries you can find data that might allow us to apply a geospatial model knowing though
that your accuracy is going to be reduced, you’re going to have greater uncertainty because
you may not have the precision that you might, fore instance, our data in Minnesota, but
I think if you were to focus on where to sample with limited resources, knowing your wastewater
inputs to the environment is always an — probably the most important input into that natural
environment, and so that would be one area to focus on. You could do, like, as we are still doing,
estimate of the influence of animal agriculture by livestock density. And again, many of these countries might have
some way of getting at that through a surrogate dataset. You don’t necessarily have to go out and sample
everything in order to build some of these models, but I think one of the most important
pieces no matter where you would build the model, is to actually field-validate the model. You make some predictions of where you think
things are loading, it’s going to vary by drug, it’s going to vary by bug, get out into
the field and see how well you’re actually predicting those amounts, and then you can
base your intervention strategies on that.>>Elaine Larson: Thank you. Dr. Boucher?>>Helen Boucher: So I just wanted to come
back to the penicillin allergy issue and just raise the consideration of implementation
because we now have these nice protocols that a number of members of the team can implement,
but a number of people across the country have encountered resistance from their hospitals
and the other places because there’s poor reimbursement for this activity, so whether
be a pharmacist, a physician, whoever on the stewardship team, there have been a number
of discussions where there’s frankly been pushback about, you know, where this fits
on the priority list and so I would just submit to the committee that we think about any ways
this might be part of an incentive, because this is the kind of thing the risks — the
people that were listed as the most likely to benefit from being delabeled are the people
most at risk of getting AMR infections, so it certainly seems like it would fit in that
umbrella.>>Elaine Larson: Thank you. Dr. Black>>Stephanie Black: Hi. Thanks to the speakers for very interesting
presentations. My question’s for Dr. Khan. It seems like with a penicillin allergy issue,
that it’s a big communication issue and as you mentioned that as soon as people get delabeled,
they may go out to another provider and tell them they’re allergic again. We’ve benefited locally in Illinois from this
development of a registry for those who are colonized or infected with extensively drug-resistant
organisms, and I just wondered if you had ever heard of any kind of efforts for any
sort of registry about penicillin allergy, anything like that.>>David Khan: Where — there is a relatively
small drug allergy registry that’s in development from a few different institutions centered
out of Mass General, but certainly not — and it’s more just, kind of, to study and gather
a little bit of information but certainly not on the scale that you’re talking about. But yeah, you know, it is a challenge and
one of the things that we’ve developed a number of different tools to try address this, you
know, what we call relabel and ranging from education, contacting the patient, et cetera,
but I think one of the low tech solutions is we give these patients a laminated card
that, you know, kind of, lists whatever drug allergies they have and then in bold font,
you know, about this big it says “I am not allergic to penicillin,” and then they can
show that to their pharmacy or wherever because, you know, we have a relatively closed system
within our county hospital, but even some of them are going outside so that’s one strategy. But there’s a lot of work that needs to be
done.>>Elaine Larson: Dr. Ginocchio.>>Christine Ginocchio. This is — I’d like to thank the group — very
interesting. Going back to the penicillin allergy, this
is really quite fascinating. So you’ve instituted this program and your
hospital where you have your pharmacist go and, of course, we’ve talked a lot about what’s
cost-benefit ratio and it’s hard to measure maybe the long-term outcomes, but in your
situation considering the pharmacist’s time, the reimbursement, how many cases did you
actually delabel and do you have any data on the impact of moving forward with antibiotic
selection and the cost related to that?>>David Khan: So we’ve been doing this program
since the last four years — well, actually, since November of 2014 actually. Yes, and we’ve delabeled — about 900 patients
we’ve been able to delabel. We — cost is real tricky, and part of the
reason is you know if we want to, you know, avoid using Vancomycin or fluoroquinolones
and things like that that’s great, but it doesn’t save much costs because those are
dirt cheap as well. So we have instituted a program for aztreonam
use and in our institution, like, probably like a lot of institutions, the majority of
aztreonam use is for penicillin allergic patients and so we now have as part of the order set
when someone orders aztreonam, it automatically, kind of, gives a penicillin allergy consultation
and we have shown by targeting aztreonam, you can save money quite quickly. So it’s been a little difficult to show cost
savings, that being said, other institutions have been able to show cost savings by instituting
some type of penicillin allergy evaluation program. But you’re right in terms of, like, from an
economic model, we started our program based on a, kind of, a quality-improvement Medicaid
waiver fund that came through. It lasted a year, and the institution said,
well anybody we hire out of this, we’re going to keep on, and they’ve seen value in what
we do but, you know, pharmacists aren’t cheap labor and it’s an expensive process, but it’s,
you know, this commitment that it’s doing the right thing and globally it will save
money but we’ve had a hard time showing it. We’re going to need, you know, we’re looking
at that, you know, years have data to see can we show economic impact, but it can be
a tough sell.>>Elaine Larson: Okay Dr. Plummer and then
Ramanan. So I’d like to echo thanks to all panel speakers
— great panel session. So my question’s for Dr. Cray. So Dr. Cray, your data set, and maybe you’re
not far enough yet to know, but you brought up this idea of these environmental impacts
resetting the AMR in the environment and so I was interested in knowing in your data set
thereof, you know, immediately post-environmental flooding, did you see consistency — you demonstrate
some data but across the board, did all samples look reset or was that not the case?>>Paula Fedorka-Cray: Yeah. I actually have another slide and, in fact,
it’s not the case which I find fascinating and it actually reinforces a study we did
when I was with another group in Athens, when I was at USDA and we looked at the Coney river
water basin. In — we in fact looked at salmonella, e-coli,
campy, and enterococcus from certain locations; Dr. Rick Miners [spelled phonetically] and
myself and you could see that there really wasn’t much movement between some of them,
where others moved more. And I think that what, you know, what we’re
waiting for is we sampled monthly and so we can see where there are changes after a weather
event, and I think that when we get down to it, those will be the more significant days
for changes that will have occurred. What I really want to see is after another
30-mile flood up the river, what’s going to happen, and maybe it takes something more
catastrophic than just regular environmental movement for rain or, you know, freezing snow,
something like that.>>Elaine Larson: Ramanan.>>Ramanan Laxminarayan: Yes, thanks for letting
me ask a second question. So back on the penicillin allergy, I was just
remembering we did this for a paper a few years ago, and looking at where the litigation
was a reason why people you know, sort of, more cautious on the side of, you know, flagging
someone with a penicillin allergy even if they’re not. And I recall at the time that, you know, when
you look at the literature, most of these happen to be settled out of court, and the
only one which was, like, a very prominent case was actually of a bull in Oklahoma call
Ferdinand who got 5 million units of penicillin and died, and then the owners sued the veterinarian
and so — it’s one hell of a story, yes — so the actual — the actual only case which actually
went to trial that we could find was this Ferdinand the Bull from Oklahoma. But are you aware of this being something
that people actually sue if they — or that people are worried about lawsuits which is
why they’re cautious against using penicillin?>>David Khan: Yeah, certainly there’s — there
is some degree of worry about this in and that even translates to cross-reactivity meaning,
you know, if you’re allergic to penicillin what about giving a cephalosporin, what about
giving a carbapenem, and there have been some lawsuits about these things as well in like
you indicated, you know, a lot of it is, kind of, settled out of court so there’s not a
lot of literature on that, but it’s something that goes through a lot of people’s minds
still and I — but I think most of it is, you know, we don’t have good access to that
literature from the legal perspective but it’s a concern.>>Elaine Larson: Dr. Marty?>>Aileen Marty: Hi. Thank you. This is for Dr. Khan. So I agree with you in almost everything you
said but you see there is a problem every time that we talk about these things is that
people tend to become too cavalier. I myself had a full-blown anaphylactic reaction
and had to be resuscitated when I was given penicillin 30 minutes after being given antihistamine
because the doctor said, “Oh, no, that’s BS you know, I’m sure it’s fine,” okay, so I’m
concerned, particularly as one of those rare people that’s at risk. Do you have any comments about this?>>David Khan: Yeah, I think we do need to
take this seriously and this is where, you know, the how do you evaluate patients, what’s
the safest way to do this, and clearly there’s been a long-established methodology for using
penicillin allergy skin testing with a full set of reagents, and that is likely the most
safe method used especially for patients with histories of anaphylaxis. Unfortunately those extra reagents are not
commercially available and that’s what we’re hoping that there’s a company that’s brought
this to the FDA and we’re hoping that will go through, but now there’s a movement, partly
for the reasons that have come up that, you know, well hardly anyone’s really allergic
on testing and why not just skip the penicillin skin testing and just give people penicillin. Do what’s called direct challenges, and there’s
been a number of studies that have shown that in — when you select patients properly, that
that seems to be safe, but again, you know, it’s going to take a lot of numbers before
you’re going to cause harm to that one patient where you miss and, you know, you probably
shouldn’t have done the oral challenge. So this is where, you know, things are moving
forward. Really trying to identify who are the low-risk
patients that may benefit from, you know, doing oral challenge which may not even necessarily
involve an allergist, versus requiring skin testing ahead of time, and this is an area
I would say of debate and, you know, we’re working actually right now on our guidelines
for a drug allergy and, you know, trying to figure out the wording about this, but your
point’s well-taken.>>Elaine Larson: Thank you. We’ll finish off this panel. I just wanted to note that it seems to me
there’s a huge, sort of, overall theme here of interconnectedness. So Dr. Cray emphasized the fact that from
the One Health Perspective, the continuum, we’re all affected daily by the resistance
ecosystem, and Dr. Dhody talked about the role of museums and such repositories to look
at historically what’s happened and how that compares with what’s happening now and what
might happen in the future. Dr. Khan talked about the importance of mislabeling
of penicillin allergy in terms of the potential contribution to the development of AMR, and
Dr. Singer talked about the environmental hotspots, their interconnectedness, and how
they might be identified more easily over time with the kinds of AI and other geosystems
that we have. So thank you very much, we’ll take a short
break.>>Jomana Musmar: Thank you to all of our
presenters. We’re going to take a 10-minute stretch, bathroom
break, so please be sure to be back here by 10:50 at the latest. Thanks so much.>>Female Speaker: Produced by the U.S. Department
of Health and Human Services at taxpayer expense

Leave a Reply

Your email address will not be published. Required fields are marked *