Subscribe
Share
Share
Embed
Communicable takes on hot topics in infectious diseases and clinical microbiology. Hosted by the editors of CMI Communications, the open-access journal of ESCMID, the European Society of Clinical Microbiology & Infectious Diseases.
[00:00:00]
Thomas: Hello, and welcome back to Communicable, the podcast brought to you by CMI Communications, ESCMID's Open Access Journal, covering infectious diseases and clinical microbiology. My name is Thomas Tängdén. I'm an ID physician at Uppsala University and associate Editor at CMI Comms. I'm joined by two co-hosts, Erin McCreary, ID Pharmacist, senior Director of Infectious Diseases Strategy at UPMC in Pittsburgh in the us and also associate editor at CMI Comms.
Hi, everyone. Excited to be here
and I'm also joined by Angela Huttner, ID physician in Geneva, Switzerland, and the editor-in-Chief of CMI Comms.
Uh, thank you Toma. Thanks to all of you for letting me crash this party. We really don't need three hosts, but I absolutely did not want to miss this one.
Erin: I think there were like six of us that wanted to host this episode actually, because today we are kicking off a new series on communicable in which we will make you love [00:01:00] PKPD and Angela.
when we were outlining this was like, we're being optimistic here, but obviously by how many people wanted to host this episode. I think this is something everyone loves and if you're like, I don't love it 'cause I don't know anything about it, you will love it for sure. By the time we get through this episode with these amazing guests.
So I'm delighted to introduce our first guest, Amy Legg. Amy is Assistant Director of Pharmacy and a consultant pharmacist in infectious diseases and research in Brisbane, Australia.
Erin: Her research has investigated the nephrotoxicity of antibiotics in patients with staph aureus bacteremia, as well as the use of dosing softwares to predict the optimal dose for a patient and to minimize toxicities of antibiotics. Indeed, her area is very much precision medicine and individualized dosing, what she calls the holy Grail of clinical pharmacy.
And I'd have to agree. Amy serves as an expert group member on the Australian Therapeutic Guidelines, an advisory board member for Australian prescriber in the Hotspots program. Love that. Don't even know what that's about. but love that name. And is a member of the Leadership Committee for [00:02:00] Infectious Diseases for Advanced Pharmacy Australia.
And finally, Amy is currently an editorial fellow at. CMI, our sister journal. Amy, we are so excited to have you on communicable, welcome.
Amy: Oh, thank you so much for having me. It's a real pleasure
Thomas: and I'm really pleased to welcome our second guest, Rekha Pai. Mangalore. Rekha is an ID physician at the Alfred Hospital in Melbourne, Australia.
She's the clinician, researcher and lecturer at Monash University. Rekha completed her ID training in Australia. Her PhD on beta-lactam antibiotic TDM in critically ill patients led to the development and implementation of Victoria's first hospital, integrated beta lactam TDM service, which she now leads.
Rekha is actively engaged in translational research, advancing model informed precision dosing, MIPD, including the integration of biosensor technologies to enable bedside real-time monitoring of antibiotic concentrations. She's a board member of the Australian Society for [00:03:00] Infectious Diseases, A SID Rekha is also, one of our editorial fellows at CMI Comms. Welcome.
Rekha: Hi, and very excited to be here. I,
Erin: I can't wait to get into this episode, but first our listeners at Communicable will know that we always start these episodes with a get to know you question for our guests and our hosts. Today's Get to Know You. Angela came up with, and I love it. I think it's so interesting. So guys, what is something that other people might find ugly, but you find beautiful. Amy, why don't you get us started?
Amy: Yeah. I was thinking about this and I guess I think of so few things as ugly, but one thing that other people seem to not like that I like is rain. I love it when it rains. There's a Paul Kelly lyric that's like a chilly city, suits a troubled soul.
And I love it when the weather's moody. I love it when it's rains. I think it's really beautiful and that would be one thing that other people always seem really sad, but it's gotta rain on the weekend and I think how bloody wonderful.
Erin: That's so funny 'cause [00:04:00] you live in one of like the sunniest areas in the world, I think.
And I live in Pittsburgh, Pennsylvania, which is rainy every day. So you should come move in with me, I think is what we're saying. It rains all the time here. Rekha, how about you?
Rekha: Hmm. I can't say I love the rain, Amy. I hate my socks wet. But I found this one really hard because taste is so personal.
But, I just remembered something I did recently and we are renovating our house and I love color. So I have a room that has orange carpets, pink walls and green chairs, and a purple armchair. I love it. My family hate it. And my husband, has been vetoing the orange carpet because it hasn't gone in yet.
So, that's my office,
Erin: my whole house is white. I am a neutral girl. not that I don't love color, I'm I have like no eye for decor, so you should also move in with me and decorate my house. I appreciate that.
Rekha: I don't think you'll love that.
Erin: No, it sounds fun.
Alright, so now that we know a little bit more about our guests, it's time to dive into this series. [00:05:00] In this first episode, we're going to lay the groundwork.
So even if you are an experienced PKPD researcher or you've never heard the term, we hope you find value in kind of this level set definition background in this topic. So we're going to start with general questions. Rekha, we're gonna start with you. When you say PK and pd, what do you think are the minimum concepts a clinician should picture?
What does PK mean? What does PD mean? And then what does that include?
Rekha: So pk, which is pharmacokinetics, so pharmaco, which is drug kinetic, which is movement. So basically PK is the movement of the drug through the body and what that journey really involves.
So what the body does to the drug while the drug moves through the body. So if you're thinking of a drug that's administered, via any route, except for intravenous, you would have to think about it undergoing absorption. So that would be the first step. And then it distributes throughout the body and different [00:06:00] compartments, so plasma and all the other tissues, other organ spaces.
And that's basically the distribution. And then it gets metabolized to either it's active form or inactive form, and then it finally gets eliminated from the body. So that's really your PK and the ADME where it comes from. And there is a concentration versus time profile in all those compartments, during its journey.
So that's really, PK PD is the exciting aspect because it's the effect of the drug. So for a pathogen, it is either the -cidal or the static, effect or the, emergence of resistance, depending upon what kind of concentration of the drug is at the site. So really it's the action of the drug and the PD component for antibacterials is the MIC or the minimum inhibitory concentration of the drug.
So that's really what PD means and. There is another effect that we think about when we think about pd, and that's the toxicological effect of the drug as well, and that's on the host. So [00:07:00] that's what I think of when I think of PKPD from a very basic level, and that relationship between pk, which is the journey of the drug and pd, the effect of the drug is the PKPD index, and that underpins the efficacy of that drug.
So that's where those concepts come from.
Erin: Thanks for that background. I think that was a really good explanation. Hearing that, what do you think are the most common misconceptions when clinicians hear at PKPD and try to speak about it, but may not be versed in the space?
I think firstly, a lot of times clinicians think that they don't do PKPD, but they actually do do PKPD.
Rekha: It's pretty much literally when you're thinking about should I load someone or not? Should I extend an infusion or not? What dose should I choose and is this toxic? So I think everyone really does some amount of PKPD. I think misconception is, it's too complex. I hate it. I think we need to kind of move away from that sometimes.
the second thing I believe is that people can, we can get pretty fixated on very rigid rules. I mean, as, five [00:08:00] of us here, we know that standard dosing is not really okay for patients. But a lot of times clinicians can get fixated on standard dosing and may think that. they need to think outside the box only in an ICU setting.
Rekha: But, the truth is that standard dosing may not be applicable to even patients outside of ICU. so I think that's another a misconception where we can be pretty rigid, the way we dose patients and the way we think about, how we want to dose certain infection syndromes.
the other thing I feel is important when it comes to sort of rigidity of thinking is, PKPD indices. So, my main research area is beta-lactams. Just about five or six years ago, we thought that 50 to 70% free drug over MIC was sufficient. But now we know that we need a higher, a hundred percent time over MIC, and in fact we need higher concentrations as well, up to four times over MIC.
Rekha: So that sort of shift in thinking is important. As the patient population changes or as new data becomes available to keep up or be at least fluid [00:09:00] in accepting that there might be changes to your PKPD targets from how you might traditionally think about it. thinking that MIC is a fixed number and it's the all important number I think is, another sort of misconception.
And I think it gets a lot of weight to put on it. But the mics that we deal with in everyday life are not really the sort of gold standard MIC. So it's important to realize that it's a derivative sort of number. And then finally, Thinking that T DM is an answer to every question, and believing that, the concentration in the blood is the ultimate, whereas it really may not fully reflect the concentration at the site of infection.
So, that's another thing to be, thinking about when you have access to TDM. so those are, some of the misconceptions that we would, come across when people think about PKPD.
Thomas: Yeah, thank you so much, Rekha And we'll come back to many of these things and dig a bit deeper on, those points later.
moving to you, Amy. let's talk about the PK part of the PKPD, what the body does to the drug. [00:10:00] How would you explain this in simple terms? Like what do you think clearance, volume of distribution, halflife actually means in a clinical situation?
Amy: these are concepts of course, that we all learn about, and then I think we think we can get rid of them over time.
But these foundations are so helpful when we're thinking about what is the pharmacodynamic parameter I'm trying to achieve and what dose do I need to give to achieve that? And working out what dose to give is really understanding pharmacokinetics, as Rekha beautifully said, pharmacokinetics really describes the relationship between the dose of the drug that you're giving and then the concentration time profile that you're gonna achieve in the body.
So if we focus on the three that you mentioned, I'm gonna start with volume of distribution because that's the pharmacokinetic parameter that really predicts whether you're gonna need to use a loading dose. So that's where I like to start there. It's really a sort of, artificial parameter.
It doesn't actually describe exactly where the drug goes or, how [00:11:00] much of drug goes, different places. It's just giving you a rough idea about how much it moves out of the bloodstream. There are a couple of key things that drive a drugs volume of distribution. One is it's protein binding, so drugs that's really high protein bonding get this nice reservoir in the bloodstream and so the drug stays in the bloodstream.
We often poo poo protein binding, I think because we think it, you know, holds up the drug. It doesn't let it be free and active, but it's actually relatively useful to have a bit of protein binding. It holds onto the drug in the bloodstream. It gives the drug a bit longer half-life, so it's can be really useful but of course it does limit the amount of free drug available to work.
Angela: It is so funny the way you describe volume of distribution, because I literally, every time I think of volume of distribution, I think of like pouring some liquid into a big, like, amorphous container.
Yes. That's the volume of distribution, there aren't really great metrics for it, right?
Amy: No, totally. [00:12:00] I think sometimes I wonder that we try and make more of it than it is.
So just because your drug has a big board of distribution. You don't know. Oh, that definitely gets into the cs. You know, you've still gotta check what's its, pharma co parameter is. And I think the other thing that's really helpful with, as you say, thinking of volume of distribution really as a bucket and we need to fill it and how much do we need to give to fill it.
but really volume of distribution's a mathematical equation. if you look at how much drug you give and what the concentration is, you just derive the volume of distribution from those two parameters. So, I guess to summary of all of that is volume of distribution's really, really important for dosing because it, leads you to whether or not you're gonna need to give a loading dose.
but. it's not an absolute sort of definition of drug penetration of sometimes I just think can get a little bit mixed up if it's got a big volume of distribution. Doesn't mean it's gonna have excellent penetration into all different tissues. It probably does, but you still gotta check it [00:13:00] out.
the second thing I'll talk about is clearance, as you mentioned. So once we've filled up our bucket, clearance is of course the movement of drug out of the body, it can relate to metabolism as well, but essentially the clearance of the drug from the blood per unit time. And so that's like how much drug do you need to replace really relates to your clearance.
So clearance is the pharmacokinetic parameter that really drives your maintenance dosing and ongoing dosing. and clearance tends to come from the kidneys, but of course it can be, cleared by the liver, it can be cleared through bile. we've got esterases in our blood that break down some of the echinocandins.
So there's lots of ways that drugs can get removed and clearance is really crucial. Particularly we know patients who have renal failure. When they've got a drug that's renally cleared, we have to be thinking about giving less 'cause we're gonna have less clearance. So we need less top-ups. And then half-life is related to both of those parameters.
So it's a, mix of both the volume of distribution and the drug clearance. And it's really talking about how long does it take for the drug [00:14:00] concentrations in blood to halve. and that's really talking about how often we need to be giving our top-up doses or maintenance doses. So I think those three pharmacokinetic parameters, volume of distribution, clearance, and half-life really help us define the optimal dosing regimen for our drugs.
Thomas: Yeah. when it comes to clinical implication, you're saying that Well, the volume of distribution is the most important PK parameter when you initiate therapy, like to determine the loading dose.
Whereas for maintenance dosing, the clearance is much more important. Correct.
Amy: Yeah. I really think we've gotta be thinking clearance, really as our sort of primary parameter. , the more you play around with PKPD, you do realize that for drugs that have a long half-life, if you don't give a loading dose, it does take you a long time to get to your, apparent steady state. And that may mean that you're sort of compromising achieving a pharmaco dynamic target.
And then you've just gotta think to yourself, how crucial is it that I achieve [00:15:00] my target immediately? How much longer is it gonna be to wait? And then if it's something that you're not prepared to wait for, fluconazole is a great example. If you don't give a loading dose, you can be waiting four or five days to achieve the PK or PD target that you're looking for.
That might not always be clinically relevant, but you know, it might also similarly be clinically relevant. We see the same thing with Vancomycin one drug that's having a real moment of loading. I don't know if this is happening internationally, but in Australia we've sort of moved towards loading Clofazimine.
and it's got a really long halflife and even though you give it for months with a halflife of 70 days, if you don't load it, you may never get to the steady state that you're hoping for or the, targets you're hoping for. So, you know, volume of distribution's, important loading doses can be really important, but I think for both efficacy and toxicity, having a good understanding of drug clearance and your patient's organ function is, probably number one.
Thomas: Just a final question about the PK part. So you mentioned the protein binding [00:16:00] and, we usually consider it only the non-protein bound part, to be active, but you said there may be some advantages actually to have some protein binding. I think that's interesting
Amy: Yeah. Well if we think about, you know, Meropenem's got like 2% protein binding.
We give it two or three or four times a day. If you're in a pharmacokinetically forward place, and Ertapenem's 98% protein binding, we can give it once a day. And it's really fantastic for patients who go home on community-based therapies. So provided you're still getting the unbound fraction that you need for the drug to be efficacious, protein binding can prolong your half life.
It provides this beautiful reservoir drug in the blood, that then, you know, comes off proteins as needed. The trick. Is of course Flucloxacillin. That is a cheeky little drug that has very high protein binding and a very short half-life. so that's one sort of trap for young players, but still a fantastic drug though.
Erin: It's funny, Thomas, that you asked about that 'cause my ears also perked up when Amy said protein binding is not something people [00:17:00] get excited about. I think increasingly they will, though I think there's data coming out.
I mean, we knew ertapenem a long time ago, the classic 2014 mortality story, whether you believe that small study or not, you know, now there's data with, cefazolin and it's a very heavily protein bound antibiotic and, you know, seeing more treatment failures or potentially higher dosing needed in, patients with hypoalbuminemia, which tracks, because the SNAP dosing is 2q6, which I don't think people realize, right?
For deep invasive infections, most of us are. Capping at 2q8. So I think hypoalbuminemia is something that we need to appreciate more and is emerging as there's legit clinical care here. When I was in residency, one of my attendings, Dave Andes, a PK/PD God told me to give ertapenem one Q 12 for patients that were obese and or hypoalbuminemic .
And I was like, you're insane. But he alas is not actually. So I think, that's an important point.
Amy: Yeah. And we've certainly used Ceftriaxone Q eight for [00:18:00] hypoalbuminemia.
Erin: yeah,
Amy: because yeah, protein binding has some advantages for sure, but you've still gotta make sure your unbound fraction is adequate.
Rekha: And are you monitoring that, Amy? , Is that because you have TDM?
Amy: Yeah, I do have, I have access,
Rekha: yeah.
Amy: Rekha does as well. But yeah, we have access to a lot of T DM.
Rekha: Yeah, I think it's easier to do that when you have access to TDM really. I mean, we've been doing Q six cefazolin, as well because we have access to TDM so we know what's going on.
Now let's go to something that might seem straightforward, but it's actually quite complicated. the MIC is a cornerstone of PKPD indices, So Amy. how do you think we should interpret mics as a lab measurement? What does it tell us? What does it not tell us? And why do we have such large variability and results, not only between labs, but also within the same laboratory?
Amy: Yeah, at the moment we are asking mics to do a lot. Uh, we're asking them to give us lots of different types of information, and I think they're struggling under the weight that we're [00:19:00] putting on them. So in very basic terms, I think of the MIC as a marker of how easy or hard it is to kill the bug. A low MIC means that the bug is quite easily killed by the drug. A high MIC means that the bug is harder to get killed by the drug. In basic terms, that sounds great. Within that, of course, the lab uses the MIC to work out whether or not it's likely to have resistance. So we have, a series of mics that are grouped underneath the ECOFF, the epidemiological cutoff value, and that's where we're really expecting the bugs to be wild type and not have any sort of resistance mechanisms.
And so that's very similar to another category of MIC, which is where the MIC falls under the clinical breakpoint. And these two concepts in the first instance confuse people a little. So we have a clinical breakpoint that takes into account drug dosing, PKPD, site of infection, clinical [00:20:00] data, and that's what we think is the breakpoint for the lab to call a bug susceptible to a drug.
We can get into other terminology for SDD I, R, we can talk about that if we want to. So there's the clinical breakpoint. there's also the epidemiological cutoff value that I mentioned, which is often similar to the clinical breakpoint, but it takes into account a bit less and really just thinks about is there likely to be phenotypic resistance in this pathogen at this, MIC.
And that's the lab side of it. The reason that there's so much variability, and I'm not a microbiologist, so I will take any people jumping in on this, but at any point the MIC can be one dilution up or down. So even when it's repeated within the same laboratory, you could get a slightly different value because it could be, a different person interpreting the MIC and then between labs, different methodologies can mean that the [00:21:00] MIC slightly moves.
So all of that. Even just the concept of having these breakpoints really I think helps me to understand that MIC was not necessarily meant to be one specific value for one infection for one person. It was meant to give us an idea of whether or not the pathogen is likely to have a resistance mechanism and whether or not the standard or optimized dosing of a drug is likely to be efficient or effective.
In this scenario, what we've now done by introducing PKPD parameters is then try to get a number that we can use to work out whether or not we're achieving a target. And that's another thing that mics were not really ever meant to do. The other thing I think about is like if you take a sample, you might not capture every single bacteria that's in that infection.
You might capture part of the infection if the infections. got multiple different sources and different foci. So the one MIC that the lab gives you from that sample [00:22:00] may not represent all the mics of all the bugs that are causing that infection. And then to get more complicated, probably the MIC is moving as the bacteria is getting exposed to antibiotics over the time course of the infection.
So I think we need to be pretty pragmatic with mics. there was a wonderful paper a few years ago that was talking about MIC dose adjustment facts and fables.
And it really proposed some rules around using an mics within PKPD targets. If we've got an individual MIC for our patient, that's just giving us information about the likely bacteria. They then talked about if your MIC is above the ECOFF then using a dosing MIC.
That's twofold. and then if you are more than twofold, above the top of the ECOFF, then starting to think about a different drug altogether because then it's gonna be very hard to achieve your targets. What I clinically do in practice also relate to the likely [00:23:00] toxicity of the drug. So I'm pretty happy to be, liberal with dosing of penicillins.
I'm not as happy to be liberal with dosing of vancomycin. Not many people outside of Australia have to worry about dosing vancomycin. But because in Australia we are vancomycin lovers, we are still dosing vancomycin and we largely just use an MIC of one, which is sort of a whole different, mindset to what I've just been speaking about.
So you do have to think hard about using an individual MIC I think. My advice would be to really think about using either the clinical breakpoint or the ECOFF if you're beneath that, and just assuming that the, breakpoint will encompass anything that's under there, and particularly if the MIC is moving over time as well.
Thomas: I entirely agree with you that it's a fascinating like determinant or measure of bacteria susceptibility, but it's not the whole truth. Like
the, the paper you mentioned has challenged. I think the PKPD way of dosing, like [00:24:00] individualized dosing was always based on the MIC value.
And when we can't trust that anymore,
Amy: yeah.
Thomas: the whole concept, especially if you're working with toxic drugs, like you said.
Amy: Yeah. It's the toxicity that really limits your ability to be frugal with the MIC because once the drug's toxic, you wanna give as little as you can. And the lower the MIC usually means the less drug you can give.
So as soon as the drug's toxic, you immediately want to use as little as possible.
Angela: I get it. This could be a whole topic for a whole other, episode, but what would be the alternative? Right. I mean, I know it's a very limited metric that we're using one MIC, Apart from using several mics, you know, making the lab do all this work, that is quite probably, I'm not quite sure we have a better model.
Thomas: Yeah. What you're asking for is, more sophisticated model to, monitor bacterial response to therapy. We're actually only look at one time, point 20 hours later, which does not say anything about the bacterial [00:25:00] killing or regrowth, like how does the curve look like?
It's a very rudimentary measure. you can compare it to, I think TDM for Vancomycin, where we use trough values in most places, which just doesn't say anything about the drug exposure. And we know that the AUC, the total drug exposure is the most important factor here. So it's, it's parallel to that.
I think we're using it for practical reasons. It does say something, but especially if you have a like. unexpected response you could say. Then you probably miss a lot of information.
I think we need more advanced methods and we'll probably get it eventually. I
It, yeah.
Erin: Yeah. I agree with all the dialogue and I think, Angela, your question and point is so well taken. the mic C's not perfect, if we don't use it. What do we use? And you know, brilliant people like Thomas are trying to develop new methods, but in the meantime, the MIC is such a core component of some of our PKPD indices.
And when I say some, I mean all 'cause their ratios in the MIC is the denominator. And so we have several different [00:26:00] response indices we talk about in the PK pd. So taking one part of each right. And so free time above M-I-C-A-U-C over MIC or that area under the curve, over MIC Cmax or peak over MIC.
You know, we take a PK variable, a pd, the MIC, and we try to make it make sense. So Rekha, looking at those different exposure response indices, you know, how do you decide which one matters for a given drug class and how do you work with them knowing the limitations of, both the numerator and the denominator?
Rekha: Yep, so the PKPD index is really what we know that links the drug exposure to the bacterial killing, and that's the best we have at the moment. Right. So just for the listeners, we, talked about free time over MIC, which is how long the free drug concentration remains above the MIC of the infecting pathogens, so that.
Beta lactams, A-U-C-M-O over MIC is the total exposure, over time, and usually when it's considered to be over 24 hours, for example, fluoroquinolones [00:27:00] and then Cmax, which is the peak concentration over MIC, significant for aminoglycoside. So these indices have been developed over time.
In fact. Over decades really, because I think the first paper came out in the 1950s or 1940s by Harry Eagle, about time over MIC of penicillins. And since then, several investigations have occurred. and these indices have been developed initially over, time kill experiments, which are the static time kill in vitro experiments, and then mouse and rat model dosing studies, and then live animal studies and then human studies as well.
how do you decide? I think basically through all these, previous, experiments and time kill studies, each of these indices were developed for each of the antibacterial classes. that's how they came up with the PKPD indices. for example, when you look at beta-lactams, the time kill experiments demonstrated that.
For the duration of time over MIC, and in fact up to four times over MIC, the concentrations, they improved killing, but [00:28:00] then it plateaued off. So that's where it comes from really. when you use it for an infection syndrome, you are actually thinking, which drug am I using here and what is my infection syndrome?
And what is the likely sort of pathogen? So that's when you decide what, exposure you want and what PKPD index you wanna choose.
Angela: For me, this is like so important. ' I have to teach this too over here.
And basically, what I say is that humans are not as smart as we think we are. That these three PKPD indices are actually observations, right? Like these are antibiotics. Penicillin was not actually created by humans. It was already there. It was created by a fungus. It has its own way of killing, right?
We didn't develop these indices. We have just been able to observe them. and then once we get the observation, you know how these things kind of happen naturally. We put it into one of these three categories, correct? Yeah. Into one of these three jars. 'cause I feel like my students are often really confused.
They're like, why did we make it so complicated? Why did we develop it like this? And I always tell them, [00:29:00] look, these are just observations. nobody would design something so complicated. tell me if I'm doing something totally wrong. 'cause my feeling is there is no, intelligent design here.
This is just the way life is, the way these antibiotics work and what we are writing down when we observe them,
Rekha: that's literally it. Right? They just expose the bugs to different drug concentrations and different dosing schedules and observed an effect. So, yeah. they're already there.
Angela: These indices. It's just that we are, you know, trying to. Describe them in words to ourselves. Yeah. And also some drugs jump between different drugs.
Rekha: Exactly.
Thomas: Sometimes you have an AUC that fits better than the Cmax and other studies is still the opposite. So
Angela: yeah. what What? Anouk Muller. And Johan Mouton used to tell me was, there's all this overlap.
You know, you can't fit these drugs neatly into one of the three indices. Like, it really rarely works that well. And in fact, I think, if I'm not wrong, many of our antibiotics fall into the A UC over [00:30:00] MIC.
Rekha: Yeah.
And
Rekha: it's really the primary PKPD index, right. That is the driver. And then there are other, case in point is beta-lactam.
You can increase the concentrations to a particular limit. So it, there is still a concentration dependency there. but it's not the main thing. So yeah. So there's a primary PKPD index and then there is of course some fluidity across the indices.
Erin: Nick Iona, Frank once told me that beta-lactams were time over MIC and everything else is AUC and just keep it simple and, all the Cmax stuff.
Was really just 'cause AUC was incredibly hard to dose in the eighties and nineties. Yes. It was just incredibly hard to operationalize in the absence of, software. And so and prioritizing patient safety is really where aminoglycoside dosing and, whatnot came from. So fascinating.
Okay, so we talked about the indices, we talked about the MIC, the importance and the limitations. Rekha, how does that all come together then in terms of breakpoint setting? So another time where an MIC is a live or die [00:31:00] number, right? If it's over this number, it's resistant. If it's under it's susceptible, that's very meaningful.
People take that to the bank 'cause that's the only information they have to say, yes, I can use this drug, or no, I can't. But what's the role of PKPD target attainment in breakpoint setting and when is it not that simple? You know, when is it resistant? But we can maybe use PKPD to safely treat a resistant infection.
So breakpoint setting is, Fairly. Complex, it's a meeting of experts have realized and, they do a good systematic review of all the information that's put in front of them. And there's no sort of magical algorithm, but it's really a deep discussion on what data and evidence is ahead.
Rekha: So really it starts with the MIC distributions and the ECOFFS. You have, a database of all the MIC distributions. You have a database of the ECOFFs, which is the wild type population. So you have that information. And then you have basically a PKPD target that's already defined for a particular antibiotic.
And it [00:32:00] comes from, like we just talked about, from in vitro time kill studies, from animal models and human exposure studies, exposure response studies. So you have a PKPD target, you have all these MIC distributions. And then you have the sort of, population PK studies and multicolor simulations where, you have some amount of understanding of how various dosing regimens, behave in different clinical scenarios or, different, sort of, Creatinine, clearances, et cetera. and there's a probability of target attainment across, uh, breadth of mics. now you have the, MIC distributions, you have the simulation data, and then you have the sort of, PKPD index that you're thinking about. And then you have clinical outcome data.
So this could come from clinical trials or registration trials, or, observational studies that may have looked at clinical outcomes, success failure based on MIC, based on exposure response, safety, toxicity data. All of that is considered into setting a [00:33:00] final breakpoint. And it's just literally sitting and really, bringing expertise from, if you're thinking about EUCAST expertise in a room that would discuss, all the data from these various, sources, and then you come up with a breakpoint. So that's how a clinical breakpoint is set in a sort of a simple way. I'm explaining it in a simple way.
You just need a lot of expertise when you're actually trying to, analyze what's in front of you because there's a myriad different data points that you have to consider when you think about a break point. And it's not simple. and it's really closely linked to the dosing regimen, right?
So if you have a clinical breakpoint and you have a dosing regimen that's attached to it, and if you decide to change the dosing regimen, then the breakpoint doesn't apply anymore because you're using something different that. From what the breakpoint recommends.
Amy: I think that was just such a great point. Rekha and I, just think the other thing is the movement of breakpoints can put drugs out of business. So you know, we start by thinking, how is this drug gonna work? What's the PD indices that I'm looking for?
And if [00:34:00] we do believe that, you know, for a minute, glycosides, it's Cmax to MIC, well that's a perfect way to dose them because they have low protein binding, we give them as a quick infusion. They're cleared quickly, they don't hang around for a long time. But as we move towards different PD indices as our targets for a drug class like aminoglycosides and we think about what their pharmacokinetics are, they can't achieve the AUC to MIC in the same way that they can achieve a Cmax to MIC because of their pharmacokinetics.
So changing the PKPD target, or really the PD target can also put drugs out of business. It's a really good way of highlighting the marriage between PK and pd.
Thomas: I agree and we'll discuss more on this topic in the later episode, I do think the clinical breakpoint setting impacts clinical practice a lot more than people realize, because, like you said, Rekha embedded in the clinical breakpoint that we use a specific dosing schedule.
And if we don't, we can't use the clinical [00:35:00] breakpoint. And if they change based on some simulation on PK PD model, it's gonna tell us to not use the drug anymore despite the, there are no clinical trials, but say we shouldn't. So it's really important and we can discuss much more about this.
Rekha: Yeah. You know, when, EUCAST, created SIE for Piperacillin/tazobactam and how we shifted to using previously when it was I, the clinicians wouldn't use.
Piperacillin/tazobactam. So this is a positive example really for pseudomonas. But then when it went to SIE, now, the pseudomonal dosing has kind of set in into clinician's mindset. So they actually use extended infusions or prolonged infusions of Piperacillin/tazobactam now when they see SIE to pseudomonas. So that's an interesting sort of, you know, breakpoint shift, moving to SIE.
So I think that was one. And then, you know, the fluoroquinolone breakpoints when, they changed in CLSI in 2019, that kind of changed how people looked at ciprofloxacin and dosing and gram-negatives as well. So it can put an [00:36:00] antibiotic out of business, put it back in business.
Erin: Yeah,
Rekha: it's interesting.
Fascinating.
Angela: Hey, maybe this is a really good moment to let our listeners know that soon.
Thomas and Nav, will be hosting the leadership of eucast where they will be talking a lot about breakpoints and if people have questions, they should write to us because it's gonna be an ask Me Anything episode. So you can hold the eucast leadership to the fire, with all kinds of questions about break points and anything else you want.
Amy, we covered a lot of topics here that are all related. So how can you combine all this, in clinical practice when you're trying to optimize dosing regimen?
Thomas: How do you translate PKPD into the dose interval infusion strategy and duration, especially in specific patient groups like the critically ill patients. do you tell people at the bedside you're doing PKPD or TDM, or how do you explain this?
Amy: Well, I tend to use the phrase dose [00:37:00] optimization. So I'll quite commonly say on an ICU round, is this someone we should think about dose optimization.
And there's hardly any clinicians who would want to be accused of not practicing dose optimization because it's good for patient care. So I think about this a lot. I do think way back at the start, we talked about how PKPD sort of intimidates people, so I tend not to use that phrase so much. I really try and separate PKPD if I'm doing like academic detailing or teaching on the run.
But for example, one thing that comes up all the time, I see Australia as a real PKPD dose optimization leader. And one thing that we still get wrong, which makes me think all the time are we, is we'll have a patient on Meropenem in the ICU and they will have started on one gram eight hourly and they're not getting better.
They're worried their, presses [00:38:00] inotropes are going up and so they feel the need to increase the dose and so they in increase the dose from one gram eight hourly is a 30 minute push to two grams, eight hourly is a 30 minute push. I just, in my heart, think that's flying in the face of PKPD. And that's just a moment of us not taking these principles and using them.
I think if you really wanna use two grams, eight hourly because you really feel that this patient needs more than a gram eight hour, I absolutely can understand that, but at least give it as an extended infusion because this is a drug with very low protein binding. It's got a very short half-life.
If you've got a patient who's rapidly clearing the drug and you're still giving them eight hours to clear it, and you know that the PD target you want is an increased time of the free drug above the MIC of the pathogen, then how are you gonna achieve [00:39:00] that with your dose, with your pharmacokinetics?
You're gonna achieve that by prolonging the infusion. so I try and think all the time What have I told people? what is the PD indices? What, what do they know? Most people know that beta-lactams work by free team of MIC. So why aren't we making the dose changes at the bedside that optimize that?
And sometimes it's just a moment of not thinking, maybe it's great that no one else is thinking about it, but just all the time I feel like I'm saying give it over two hours, give it over four hours. And some of it is people wanting a real set infusion time.
So I just say start by giving the infusion over half of the dosing interval. If you can't give a continuous infusion or you're worried about continuous infusion, for some reason, give the dose over half the dosing interval. And piptaz as well is another one. you know, we do achieve quite high piperacillin concentrations.
Again, it's cleared relatively quickly. Again, I just think, you know, if we just moved to having piptaz over two hours, unless there was a [00:40:00] really good reason for giving Meropenem two grams IV as a push, there are some people who truly believe that that is a, a better dose for CNS penetration. That's a, a topic for another discussion, I guess.
But if you're just doing it for a gram negative in bloodstream or a a urosepsis, you know, prolonging that infusion is the embodiment of PKPD in practice and it will improve the time of MIC and that will improve patient care. so I just think all the time there, there are other drugs that are trickier, like amphotericin for example.
We think amphotericin probably is a Cmax to MIC drug. and it's very tempting to reduce doses or give part doses. And I think to myself, well. If we, if we're saying it's a Cmax to MIC drug, and there's some concern about giving big doses every day, the smart thing to do would be give it 48 hourly, but giving the full dose, you know, trying to optimize that Cmax to MIC ratio, post antibiotic effect.
Although, you know, for the newer, amphotericin, there's not a lot of reason to give part doses anyway. But, the [00:41:00] marriage, again, thinking about what's my PGA PD parameter? What simple things can I do at the bedside to optimize that? and sometimes we still get it wrong, even in the most basic, situations.
And the new beta-lactam, beta-lactamase inhibitors are coming, pre dose optimized. People are just learning that they're given over two hours because that's how they've come licensed and they think it's to do with stability or something, but it's really just, PKPD in practice coming through the drug development stage.
Thomas: Yeah. So what you're describing now. I also see that there's a gap between PKPD knowledge and clinical practice. we recently shifted towards using prolonged infusion always in the ICU setting here. the first proposal for me was to use it only when you had a gram-negative infection. And the ICU doctors told me, that's not practical.
We need to do exactly the same for all patients. It's not possible. And, would you agree that this is a test that we should take on together, like trying [00:42:00] to translate the PKPD knowledge We have to clinical practice and I guess we have to simplify it a lot to make people trust it or, get used to, applying this on an everyday basis.
So interesting that you say that, Thomas, because, we implemented continuous infusions in our ICU and we simplified it by making it for all comers. So there's not much thinking there. you see meropenem, you you see Piperacillin/tazobactam, you start the continuous infusion and it's been wonderful.
Rekha: The uptake is really, really great and, barely seeing any subtherapeutic concentrations now.
Everybody's optimized.
Erin: I agree completely. And I say this a lot. You have to make doing the right thing easy and yeah, we do three hour infusions of cefapime, meropenem and Piptaz.
System-wide, whole house, every patient. And it comes up a lot of like, well, if they're on dialysis and no residual urine, do they really need a prolonged infusion? Probably not. But if I'm a nurse, I'm gonna do the same thing every time for the drug. and they're already getting other stuff and we just don't modify it based on some [00:43:00] of those things because then you're not missing the patients where it really does matter in, and I think, this is like our whole world of being pharmacists, but it's almost like operationally we kind of wanna work ourselves out of a job because we wanna build these structures where these things are just done.
So Rekha, to your point, if everyone's getting continuous confusions out the gate, you don't really need to do TDM, you know it's safe, you know it's effective and like individualization is cool, but is it really adding that extra added value? And that's a whole nother debate we can get into in future episodes,
Rekha: right?
Yep. I think so. Yeah.
Erin: Alright. Well guys, this has been amazing. I think we always wrap up with, because we've covered so much in a short amount of time, and I hope our listeners are like, wow, you've really given me a lot to think about. 'cause I know you guys have, and this has been really awesome discussion. Is there anything you want to tell our audience that you haven't had the opportunity yet with this episode, or that we haven't specifically asked about?
Knowing again, listeners, there are more PKPD episodes coming where we're gonna get into all kinds of fun things like hollow fiber models and more about target attainment and all that. But for now. Amy, any final words of [00:44:00] wisdom?
Amy: Yeah, I might just say two things. One is, we're all very passionate about this.
I think we've thought about it a lot. I think about it at a very high level. But I hope the listeners understand that it's all based on a series of assumptions and it's just leading us in the right direction. And you can get really hung up on exact targets. And I just hope that what we can all understand is the principles and trying to make sure we are putting the principles into practice.
And just not to get too bogged down on the exact numbers, which I'm sure will come out later. The second thing I'll say that I should have said when I was talking about mics is when we're talking about exposures and toxicity, mics don't matter. I just find that confuses people a lot because our pharmacodynamic target for efficacy has MIC as a denominator, but our exposure targets for toxicity are just numbers.
and they're the two things that I just wanted to say. Thank you.
Rekha: That's a great point [00:45:00] because the toxicity is a action of the drug, not of the bug. So it shouldn't be a multiple of MIC at all. I agree. We don't have MIC as a, toxicity target. I wanted to just remind listeners that. it's important to put everything into the clinical context, so don't forget the patient at all.
When you're thinking about PKPD principles and numbers and targets, et cetera, there is a whole clinical syndrome in front of you, so don't forget the clinical nuance. There is, you know, so many other things going on, in terms of the site of infection, and also response. So sometimes it's not just about getting the right number, it may be a right number for that particular patient, or it may not.
So I think it's important to put the whole clinical context in the picture.
Erin: And I'll just say not to like end this by bringing us back to the beginning and putting us in some PKPD matrix. But you could argue that the MIC relationship with toxicity is the break point. And so there is an MIC in which we've said over this MIC, you cannot [00:46:00] safely treat this pathogen.
Right. And then we just manipulate that from time to time and we're more willing to give more aggressive doses for higher mics that are resistant if it's a safe drug. And so it all just goes around forever and that's why it's so cool.
Thomas: Excellent.
Thank you so much to our guest today, Amy Legg from Menzies School of Health Research and Rekha Pai Mangalore from Monash University.
Angela: And listeners note, this is not the last you will hear from Amy and Re We are very, very happy. they are going to officially switch roles. They're now going to be co-hosts with Thomas of the remaining episodes in the Love PKPD series. So please tune in to see who their next guests will be and what aspect of PKPD they'll be taking on.
In the meantime. Thanks for listening to communicable the CMI Comms podcast. This episode was hosted by Thomas Tängdén in Uppsala, Sweden. Erin McCreary in Pittsburgh, USA and a little bit me, fly on the wall. [00:47:00] All of us editors at CMI Comms ESCMID's, open Access Journal. It was edited by Katie Hostetler, O and Peer, reviewed by . Ummu Afeera Zainulabid, international Islamic University in Malaysia in Kuantan Malaysia. Theme music was composed and conducted by Joseph Meade. This episode will be citable with a written summary referenced by a DOI in the next eight weeks.
And any literature we've discussed today can be found in the show notes. You can subscribe to Communicable wherever you get your podcasts. Thanks for listening and helping CMI comms and ESCMID Move the conversation in ID clinical microbiology and PKPD further along.