MIC vs. Resistance Mechanisms: What is Needed to Treat MDRO Gram-Negative Infections?

In today's session, our experts will explore the drugs available for treating MDRO gram-negative infections, and outline the challenges and opportunities associated with MDRO gram-negative diagnostics.

Please submit your questions for the Q&A session at any time during the webinar using the button to the left of your screen. And without further delay, I'd like to hand over to our speakers.

[00:01:30] All right, today I'll be talking about new drugs for treating multi-drug resistant gram-negative infections. I really appreciate the opportunity to be here on this webinar with you as well as with these other speakers who I greatly respect, and thank you for this invitation. So we've come a long way. So over the last 10 years, since 2014, we've had a number of new antimicrobials, especially [00:02:00] antibiotics that have been approved. So before 2014 when we've had to deal with carbapenem resistant enterobacterales or highly resistant pseudomonas aeruginosa, we've had very limited options. So colistin or polymyxin B was our main core agent, plus or minus high dose carbapenem or tigecycline, or some other combination. But regardless, this core drug of [00:02:30] colistin or polymyxin B has been suffice it to say, less than ideal. The toxicities, the efficacy, the effectiveness of this agent in treating these severely ill patients has been extremely poor.

And so since 2014, thankfully we've had a number of new agents that have been approved, and that's what I'll be going through today to help set the stage to understand these new agents and where they fit in our treatment paradigm. So we'll be talking about [00:03:00] ceftolozane-tazobactam, ceftazidime-avibactam, meropenem-vaborbactam, those are some of the novel beta-lactam, beta-lactamase inhibitor combinations. Plazomicin, a novel aminoglycoside, eravacycline, a novel tetracycline, and then some more beta-lactams that have come out subsequently, imipenem-relebactam, cefiderocol, and then our latest beta-lactam beta-lactamase inhibitor, sulbactam-durlobactam. So I think we'll have a lot to discuss and it can be relatively confusing because the number of [00:03:30] agents, but I think the fact that there are so many is a good thing. So we'll sort of walk through this.

And the first thing I want to touch on is that these agents weren't just sort of stumbled upon. This is very precise in the manner that these agents and these beta-lactamase inhibitors were crafted and understanding their structure-activity relationship and how we can utilize them to target specific organisms, specific resistance mechanisms and essentially overcome [00:04:00] that. So I won't walk through every single one of these agents, but I just want to note just a few just to help us understand what we mean by the precision of these drug design and drug development. So for example, ceftolozane, this is an agent that is very similar to a beta-lactam antibiotic that we know, anti-pseudomonal beta-lactam, which is ceftazidime, but has been enhanced in its structure and its activity, and it's been able to show us that it's stabilized versus AmpC production and overproduction [00:04:30] by pseudomonas. And that's part of the reason that this is such an excellent agent for multi-drug resistant pseudomonas aeruginosa because that stability that it provides against specific resistance mechanisms that pseudomonas is known to produce.

Another good example is vaborbactam. So this is a boronic acid beta-lactamase inhibitor, and specifically this works against KPC. It's an excellent, excellent beta-lactamase inhibitor specifically to help prevent KPC from [00:05:00] causing its havoc on our beta-lactams and restores the activity of the paired beta-lactam agent, which in this case is meropenem. And there's other advantages, with vaborbactam, it's thought to have this reversible covalent bond. So then it's not broken down by the KPC enzyme, but it potentially has the ability to be recycled and inhibit more KPC carbapenemase to elicit its activity and protect our meropenem. There are other agents like [00:05:30] plazomicin, here on the side. This is an aminoglycoside, but it's an analog of an aminoglycoside where it's been sort of enhanced in its structure and that provides stability against aminoglycoside modifying enzymes, which is the primary mechanism of resistance that we see with aminoglycosides.

And there's a number of other agents here, cefiderocol, I think we know about how it's sort of a mishmash of ceftazidime and cefepime as well as it has this iron binding site that [00:06:00] allows it to sort of sneak through the outer membrane of gram-negative bacteria and to allow it to elicit its activity at the penicillin-binding proteins. So all of these agents, whether they're the direct agent like the beta-lactam or the tetracycline, or it's the sort of companion agent, the beta-lactamase inhibitors, all of these have very specific roles and niches to be able to overcome resistance mechanisms. So they all have some very specific target in [00:06:30] how we use this. The in-vitro activity I think helps set the stage to understand where these agents are going to be used from a clinical standpoint, but just to note that just because something may have some in-vitro activity doesn't necessarily equate that to being clinically effective or at least having the data in the clinical realm to understand its effectiveness.

So just to walk through this, I'll walk through each agent and just sort of point out very roughly, and [00:07:00] it is oversimplification, but I think it just helps us get some context to where these different agents fit in. So in chronological order, ceftolozane/tazobactam, this is our pseudomonas agent. It's not helpful or active versus CRE or Acinetobacter, but this is an excellent agent against multi-drug resistant pseudomonas. Ceftazidime/avibactam, this is something that has a couple rules. So this was our first KPC agent [00:07:30] or anti-KPC agent that was a beta-lactam or beta-lactam/beta-lactamase inhibitor. So this sort of revolutionized how we manage patients with KPC producing infections. So not only is it active against CRE, especially KPC, but it has activity against OXA-48 like enzymes, which is an emerging carbapenemases that we're seeing in the United States and obviously globally. It also has activity against multidrug-resistant pseudomonas. So it has a couple roles that we can play when we think about [00:08:00] MDRO gram-negatives.

Meropenem/vaborbactam, this is very specific. So this is very specific to KPC. It doesn't have activity against metallo-beta-lactamase, carbapenamases or OXA-48 like enzyme carbapenamases. This is very specific to a KPC producing enterobacterales. Plazomicin and eravacycline, these are not beta-lactams, so thankfully they don't have to deal with the issues of different carbapenamases or beta-lactamase [00:08:30] that are the common mode of resistance that we see with a lot of these gram-negative MDROs. Just to note though, while it's not subject to hydrolysis by beta-lactamase, there are some resistance mechanisms, for example with plazomicin, that might be co-expressed with NDMs, for example. So just because it's not subject to hydrolysis by an NDM, you just have to be careful that we have to think about [00:09:00] is there a co-expression of other resistance mechanisms against plazomicin, for example.

But with that sort of understood, these provide a special role for being able to treat certain infections given its activity against CRE. And then eravacyline also has some activity against acinetobacter, but very limited clinical data and we'll touch on that in a bit. Imipenem/relebactam, this is in some ways sort of similar to ceftazidime/avibactam where it has CRE activity, [00:09:30] but specifically KPC as well as multi-drug resistant pseudomonas. One notable difference though is it does not have activity and relebactam is not helpful in the context of OXA-48 like enzyme producers. Cefiderocol, like I said, has a very unique resistance mechanism of action, and that provides very broad spectrum activity against many different multi-drug resistant-gram-negative. And then sulbactam is a very interesting contrast to that. So while cefiderocol has [00:10:00] broad spectrum activity against CRE and pseudomonas and acinetobacter and even stenotrophomonas, sulbactam/durlobactam, is very specific to acinetobacter. So this provides a much needed option when we're thinking about carbapenem-resistant acinetobacter baumannii, which we have very limited options for. And we'll touch a little bit more on what we actually would think about clinically.

So just to note the FDA approved indications, as we would expect when we're thinking about [00:10:30] antibiotic agents that have activity against gram-negatives, these are the typical indications that we always see. UTIs, pneumonia, particularly hospital acquired or ventilator associated pneumonia and intraabdominal infections, because these are predominantly driven by gram-negative bacteria in terms of the etiologies. Our initial agents, ceftolozane-tazobactam and ceftazidime-avibactam, these have indications actually across all the different disease states that you see here. While cefiderocol is specific [00:11:00] to UTIs and VAP, or hospital acquired pneumonia and ventilator associated pneumonia. Eravacycline, this specifically only has an indication for intra-abdominal infections. And this is notable because the interesting part about the pharmacokinetics of these tetracyclines, including eravacycline, is it has a very wide volume of distribution, which essentially means that it has great tissue concentrations, but it doesn't have good sort water compartment concentrations, meaning like [00:11:30] blood concentrations or urinary concentrations. So this provides a great option for "tissue-based infections" like intra-abdominal infections, and that's specifically what it is FDA approved for.

Meropenem-vaborbactam, this only technically has an indication for UTIs as well as plazomicin. And then sulbactam-durlobactam, as I mentioned, this is a very focused agent in terms of its pathogen activity with acinetobacter. And so this has only been studied and it's specifically indicated for hospital acquired pneumonia [00:12:00] and ventilator associated pneumonia, which makes sense in terms of thinking about what are the most common diseases that we see caused by acinetobacter. And so just a little bit of a sampling. So we've sort of talked about this in-vitro activity, and that kind of gives us an idea of where we might position these agents in our treatment paradigm. We talked about the FDA indication just to understand the official FDA approval, but what's the clinical data look like in terms of how we actually use this and where it's role might be? [00:12:30] So just as a quick sampling just to touch upon each of these different agents, ceftolozane/tazobactam, we've looked at this early on and there was this great study that had over 20 centers and over 200 patients, specifically looking at caftolozane/tazobactam's treatment for multi-drug resistant pseudomonas infections.

And what we saw was, although this wasn't a comparative study, we saw that early initiation had an association with better survival. I think that speaks to its great activity against multi-drug resistant pseudomonas and especially the lack [00:13:00] of options that we've had previously. The second study, which is a focus for ceftazidime/avibactam, this was a prospective multi-center study or otherwise known as the CRACKLE study by the Antibacterial Resistance Leadership Group. And this looked at ceftazidime/avibactam versus colistin, and these were mostly patients with bloodstream infections and respiratory infections caused by carbapenem-resistant Enterobacterales. And what we saw was that there was lower mortality with ceftazidime/avibactam with 9% versus [00:13:30] 32% in the colistin arm respectively. So we can see how we've known for so long that colistin not only is highly toxic, but it's just not that effective in treating these severely ill patients. For meropenem/vaborbactam, which is at the bottom of your slide, actually the company sponsored trial, randomized clinical trial, phase three multinational randomized clinical trial was done, the TANGO II trial.

And this looks specifically at patients with CRE infections. So this was meropenem/vaborbactam [00:14:00] versus best available therapy, and we saw that meropenem/vaborbactam had higher cure rates, lower mortality, and lower AKI, because a lot of that best available therapy was colistin based therapy, which we know carries high rates of nephrotoxicity. For plazomicin, along with the registration trial that got its FDA approval and its indication for UTIs, there was the CARE trial, which was a multicenter, randomized controlled trial, small number of patients, [00:14:30] only 39, which sort of speaks to the challenges of being able to do these clinical trials for drug resistant infections. But this looks at plazomicin versus colistin as part of combination therapy specifically for serious CRE infections. And the mortality and the complications were generally lower with plazomicin at 24% for plazomicin versus 50% in colistin.

So just another sort of ding against colistin and its ineffectiveness for treating these patients and how these novel agents [00:15:00] really have an emerging role. For imipenem/relebactam, along with the registration trial, there was a phase three randomized control trial, hospitalized patients looking at multiple different disease states, pneumonias, intraabdominal infections, UTI, and mostly it was caused by pseudomonas aeruginosa, which is not surprising being a predominant hospital-acquired pathogen. But this study specifically looked at imipenem/relebactam versus colistin plus imipenem for patients with imipenem [00:15:30] non-susceptible infections. And what we saw was that imipenem/relebactam had lower mortality, lower nephrotoxicity versus the colistin imipenem arm. And so just once again, the same theme of these novel agents providing better effectiveness relative to this colistin based therapy, which was our prior standard of care before 2014 and these novel agents emerged. And then our last two just to quickly sample on, is [00:16:00] the CREDIBLE CR trial that was done for cefiderocol versus best available therapy for serious infections caused by carbapenem-resistant gram-negative bacteria.

And this is the phase three pathogen focused open-label, randomized multicenter study. This was done over in 95 hospitals across 16 countries, and they looked at hospital acquired pneumonia, bloodstream infections or sepsis and complicated UTIs, and most of the pathogens were acinetobacter baumannii, as well as klebsiella pneumonia. [00:16:30] And what we saw here was, in general there was similar efficacy, but one of the sort of unfortunate things that we saw was these numerically higher deaths in the cefiderocol group, and this was primarily driven by the acinetobacter subgroup. So that caused a lot of controversy and thought as to where cefiderocol's role is clinically, although it has great activity against the broad spectrum of multi-drug resistant gram-negatives. Well, what's its role clinically, especially in the context of acinetobacter [00:17:00] given these findings from the CREDIBLE CR clinical trial. And now on the other hand, we have sulbactam-durlobactam, which was recently approved in our latest novel agent, and this was evaluated versus colistin in patients with serious acinetobacter baumannii complex infections.

And so this was looked at in a multicenter, randomized active control non-inferiority trial or the ATTACK trial. And so this was done across 59 sites in 16 [00:17:30] countries. And so they had patients with pneumonia, various different types of pneumonia as well as bloodstream infections. And what we saw was that sulbactam-durlobactam had non-inferiority, and so it met its primary FDA endpoint and it also had lower mortality versus colistin at 19% versus 32% as well as lower rates of nephrotoxicity. So once again, we see that similar theme of our novel agents generally being better than colistin [00:18:00] in terms of efficacy as well as safety outcomes. And so the question now is how do you take all this clinical data that we've seen across all these different novel agents and these clinical trials that have been done, which is a very sort of welcome thing that we've seen over the years, and put that together for clinicians to use.

And I think the IDSA antimicrobial resistant treatment guidance document is a fantastic guidance document by some experts in the field that [00:18:30] really help us digest all this clinical information along with knowing how the agents work, where these agents are needed in terms of systemic antimicrobial resistant infections, and how we sort of position them in terms of preferred versus alternative therapy. And so I won't go through this in detail, but just to note that our CRE agents, it's not just, “Hey, what do we use to treat CRE?” But we have options for that if you don't know the mechanism. But if you do start to be able to [00:19:00] start to break down what is the mechanism of that carbapenem resistance, then you see that the options change. So if we're dealing with the KPC producer, we have a few options of ceftazidime/avibactam, meropenem/vaborbactam and imipenem/relebactam, very consistent with the activity that I showed you earlier.

And then we have some alternative agents including cefiderocol and then our tigecyline agents, and just notably that these tigecyline agents are not for bloodstream infections or UTI in terms of its recommendation [00:19:30] because of that volume of distribution issue that I mentioned where these great levels in the tissue, but not in sort of our water compartments like vasculature or in the urinary term. But if you look at other mechanisms of carbapenem resistance like metallo-beta-lactamase, especially NDM and OXA-48, our options change because as I mentioned in the very beginning, these agents are specific not just to the different bacteria that they treat, but to the different resistance mechanisms [00:20:00] that they can help overcome. So OXA-48, for example, you don't see meropenem/vaborbactam and you don't see imipenem/relebactam, and that's because the only novel beta-lactamase inhibitor that we have that provides protection against OXA-48 is our avibactam.

So really it's a ceftazidime/avibactam and then our broad spectrum agent of cefiderocol that are preferred options for OX-48 producers, given the limited options that our other agents provide. [00:20:30] Then for difficult to treat pseudomonas, you see that we have our primary agent of ceftolozane/tazobactam and then our other agents as well that have activity against multi-drug resistant-gram-negatives. The last thing I just want to point out is for carbapenem resistant acinetobacter baumannii, this is a very challenging infection to treat. Up until this point, and these guidelines were released before sulbactam/durlobactam was approved, you can see that [00:21:00] it's much more challenging because we've had a lack of agents that have become commercially available that have activity against drug resistant acinetobacter. And so what's recommended currently is high dose ampicillin sulbactam specifically for that sulbacta- portion plus some other agent. And what that agent is can sort of vary based on patient specific factors.

But the sort of question is now that sulbactam/durlobactam has been approved in future iterations of the guidance, given its [00:21:30] high in-vitro activity, favorable PKPD profile as well as the clinical data that we saw in the ATTACK trial, is there going to be some sort of interchange of sulbactam/durlobactam versus this Hail Mary approach of just trying high dose sulbactam with amp-sulbactam as part of a combination regimen to treat these patients? So we'll see in the future sort of where sulbactam/durlobactam is placed, but it's clearly a very promising agent given all [00:22:00] the data that we've just touched upon. So where do we go from here now knowing that we've had a number of novel agents approved recently, and this is a very welcome advancement. The question is what do we do from here? Well, one is I don't think we can sort of stop and be satisfied with this armamentarium that we have. Resistance continues to emerge. The distribution of different carbapenem resistance mechanisms, for example, is continuing to change [00:22:30] in the United States and in different countries.

And so we need more agents to help us fill these roles in treating these multi-drug resistant-gram-negatives. And we have to keep in mind that the agents that we have and even the future agents that are in the pipeline, these are targeted for specific bacteria and they help us overcome very distinct resistance mechanisms. So it's not just a one size fits all in treating MDRO gram-negatives, but we have to think about where these agents fit in our treatment paradigm and especially in [00:23:00] the context of specific resistance mechanisms. And the last thing I'll say is to sort of make this all work for the patient at the bedside, it takes very strong collaboration between the clinicians and also our clinical microbiology laboratories to understand the sort of intersection of our precision diagnostics, to understand these mechanisms of resistance as well as the therapeutics to make sure that they match along with what we're dealing with from a resistance standpoint. And with that, I will close. And once again, I just want to say thank you for the invitation for [00:23:30] being here.

Well, thank you so much for the opportunity to talk to you today about a topic that's near and dear to my heart. So how do clinicians use antibiotic resistance information to treat infections? And so leading on to MIC versus resistance mechanism, what do we need? So again, the problem statement as proposed was that phenotypic antimicrobial susceptibility testing is the gold standard for treatment [00:24:00] decisions, but the problem is emerging antimicrobial resistance and new antimicrobials for gram-negative bacteria are needed, challenging the conventional wisdom that the MIC is all that's needed.

So how are we going to tackle these new challenges and how did diagnostics play a role here? So I'm just going to give you a recent case. I think it's always helpful when you're thinking about how do doctors think about this, to actually present a case. So this was a case that was presented very recently at [00:24:30] our hospital, and it's a 2-year-old with AML, unfortunately, who is hospital day 21 after getting inductive chemotherapy and the patient was neutropenic, so no immune system. During her stay and in her workup, she had been to several hospitals around the state but not traveled outside the US.

And on the day of presentation, she was febrile tachycardic. The team had [00:25:00] her on cefepime, which is very standard in our hospital for neutropenic fever. And a blood culture became positive for gram-negative rods. What happened next is that about an hour and a half later, she had a rapid molecular diagnostic that came across with E. Coli showing that it was CTXM and OXA-48 detected. So at that point, we immediately moved to change her therapy [00:25:30] to cefepime being discontinued, and ceftaz-avi being started because that is active against OXA-48 and cefepime would not necessarily be active or ideal therapy, especially for a neutropenic with an OXA-48.

Days later, two days later, her full susceptibility profile came back and again, that showed that it was ceftazidime-avibactam susceptible and that that therapy was correct. We listed [00:26:00] cefepime as resistant, and that is because that's a rule in our hospital for OXA-48, we actually flip resistance regardless of the MIC. So what does a clinician really need to know to successfully treat a bacterial infection? We need to know the source of infection, obviously. Is it the lung? Do you have enough drug to get into the lung? Is it the CNS? Do you have enough drug dosing to get into the CNS?

And what kind of organisms would cause infection in the lung [00:26:30] or CNS or wherever your infection is? What type of antimicrobials work to kill the most predominant type of organisms at that source of infection? And so if it's say, a gut infection or a urinary tract infection, you definitely want drugs that will target things like E. Coli. What type and dose and route of antimicrobial will be sufficient to kill the bacteria? Really understanding some of the pharmacotherapy and pharmacokinetics. And then [00:27:00] lastly, is the bacteria susceptible to that antibiotic in-vitro or does it have acquired or developed resistance?

So this is a schematic and it's from an old paper by one of my mentors, Rob Owens, who's a PharmD, in how complicated choosing an antimicrobial actually is. And so I think it's really helpful even though this is sort of [00:27:30] historic in a way, from 2009, you have to know a lot of stuff and there's a lot of different factors that go into how do you choose an antimicrobial and only at one sort of lower point do the culture results actually come into play. But I think it's really important that that helps you to decide, your culture results help you refine and choose antibiotics. The problem is as we're seeing more and more [00:28:00] resistance, this may get harder and harder, especially upfront. When we talk about really drug resistant organisms or carbapenem resistant organisms, I think it's a bit different than your average E. Coli.

And I think there's a lot of reasons why. And it's because with the advent or with the emergence of antibiotic resistance, we've had to [00:28:30] develop a lot of new agents that are active against carbapenem-resistant organisms. And so in the early 2010s, 2015s, we didn't really have any new drugs and it was very hard to treat these. Thankfully there weren't as many of them around. But as they've continued to emerge, new drugs have come to market that really target these organisms specifically and target specific resistance mechanisms. What this data is showing is from a lot of [00:29:00] different kind of large databases across the US from 2000 to 2017, it shows across the board we're getting better in terms of 30 day mortality at treating carbapenem-resistant organisms. And so that's in part shown there on the top graph that we're starting to use some of the novel beta-lactam, beta-lactamase inhibitors and getting away from drugs that we know don't work very well like colistin.

The problem is [00:29:30] in the 2010s to 2018, if you were practicing in the United States and you saw carbapenem-resistant and enterobacterales, you could be pretty sure that it was either not producing a carbapenamase or it had KPC. The problem is since basically early pandemic days in 2020 on, we are seeing an emergence of metallo carbapenemase in the US in enterobacterales, as well as seeing increasing number of [00:30:00] OXA-48-like enzymes, which was the case that I just showed you of a patient that had not left the state. And so obviously acquired that somewhat locally or regionally. This is data from JMI Labs or Helio Sader, showing susceptibility from 27,000 enterobacterales across 74 medical centers from 2019 to 2021. And it's showing that for the carbapenemase producers or the CREs, [00:30:30] carbapenem resistant organisms, KPC is actually going down and MBLs are going up, so metallo-beta-lactamases are going up.

The important thing to recognize is that a lot of the new agents are very specifically targeting some of these carbapenamase. And so ceftaz-avi, with or without aztreonam, is actually one of the mainstays of how you treat a metallo-carbapenemase, so an NDM. So as we see NDMs emerge, [00:31:00] it's helpful to know because that would be the agent that might be active or a combination ceftazidime-avibactam plus aztreonam might be a good guess at being active. Meropenem-vaborbactam, imipenem/relebactam, and sulbactam-durlobactam all have targeted novel beta-lactamase inhibitors that are targeting specific beta-lactamase, class A, class B, class C, class D. [00:31:30] It sounds a lot like alphabet soup, but actually it's really, really important that people are up on these different enzymes because this is the activity across these. And so when I'm thinking about how do I target an OXA-48, I need to recognize that in an enterobacterales like the E. Coli that I showed, ceftazidime-avibactam is really my only go-to across as empiric [00:32:00] therapy, especially when I know there's a CTX-M there.

Could also argue that cefiderocol is active and could be used, but again, we wouldn't have known that that patient had a carbapenamase unless we'd had that rapid diagnostic that called out that carbapenamase. We know that this is really, really critical and it goes into the thinking as an infectious disease doctor that when you've got septic shock or sepsis at all, the sicker you are, the [00:32:30] more important it is that you get empiric therapy right. So meaning that antibiotic that you prescribe as soon as you know the patient has an infection. So this is some data from some outcomes of 390 patients with gram-negative bacteremia, and again, the sicker they were, sort of their bloodstream infection score going up and the probability of 28 day mortality, you can see that appropriate therapy has an increasing effect the sicker the patient, with [00:33:00] the higher score being sicker patients, so appropriate versus inappropriate therapy. So the sooner you get that neutropenic patient on active therapy against an OXA-48, the better they will probably do.

We are heavily reliant on susceptibility testing and antimicrobial decision-making. So again, back to sort of serious infection, I take a swing at what I think this patient should need, and again, cefepime would be standard for neutropenic therapy, neutropenic fever, at our institution. And so that's what we started [00:33:30] the patient on. And then we wait 48 to 72 hours and this is sort of what was standard or is standard, until we get culture data back. Whether or not the patient's condition improves or doesn't improve can help you further refine your antimicrobials. But again, once you've got susceptibilities back, you try to tailor to sort of the least amount of collateral damage in the most targeted therapy possible. What I would argue [00:34:00] is that resistance markers, especially with multi-drug resistant organisms and carbapenem producing organisms can allow for tailoring in this space, especially in a patient who's bacteremic.

We know that this is impacting outcomes. This is some nice work from Mike Satlin looking at the use of a rapid diagnostic, a rapid molecular diagnostic, very similar to the one that we were using, showing that KPC PCR [00:34:30] from a gram-negative rod bacteremia, actually for the patients that ended up having CRE, impacted outcomes, it also impacted mortality for 30 day survival across these patients. So patients that had the KPC PCR got therapy sooner that targeted KPC and had a better outcome and a higher survival actually at 30 day lookback. There was no difference at earlier [00:35:00] time point.

What's important here is also showing that the non-KPC CRE producing bacteremia didn't really make a difference, which is what we would hypothesize. And again, if you could have a crystal ball and just say, I know that person's going to have KPC in their bloodstream, then maybe you could start meropenem-vaborbactam or an agent that targets KPC. The problem is, it's very, very hard to know when, and if we just started prescribing [00:35:30] meropenem-vaborbactam to everyone, we would start to see unfortunately emergence of resistance to meropenem-vaborbactam. So we want to reserve those drugs for the patients that need it the most.

Well, okay, Amy, fine. So you're talking all about all this carbapenemase testing, but isn't it just the MIC that matters? And what about all this stuff about ESBLs? Well, I actually think it's slightly different when we talk about ESBLs. And so I'm going to ask [00:36:00] you to follow me a little bit in this nuance. I think that we've got targets that target CTX-M, which is the most common enzyme for several organisms, E. Coli, kleb-pneumo, and proteus. It's the most common enzyme that causes an isolate to become resistant to cephalosporins or extended spectrum cephalosporins and potentially pip-tazo.

So why shouldn't we be doing ESBL testing? [00:36:30] Well, what I would say is the MIC does matter here, but the bigger issue is that there's a lot of mechanisms that can work together to synergize and cause ESBL phenotype. And when there's multiple mechanisms present, an ESBL test in-vitro phenotypically does not perform very well. It's also not very practical. And so here I would argue that [00:37:00] using ceftriaxone as a marker of potentially ESBL production is an okay surrogate and also to really be thoughtful and if the MIC is low enough, we know that patients actually have reasonable outcomes even in ceftriaxone resistant non-susceptible. So I don't feel the same way about carbapenemase testing as I do for ESBL testing because so many of the therapies also just target those enzymes specifically, [00:37:30] whereas for ESBLs, it's meropenem across the board regardless of the enzyme that you have.

I think again to the ESBL point, what I used to go around saying is you should do carbapenemase testing, general carbapenemase testing like the M-CIM where you take the disc and you put it in with your isolate, you take a meropenem disc and you put it in with the isolate. If there's a carbapenemase present, it will degrade all [00:38:00] the meropenem, and then when you put it on a lawn of E. Coli, there won't be any meropenem to cause a zone of inhibition compared to this other isolate that has a nice big zone of inhibition because that one, that isolate didn't have carbapenem resistance. Unfortunately, this would work if all the isolates had just KPC, which was probably true in 2015 to 2018 like we were looking at from that data. But the problem [00:38:30] with that test is it doesn't tell you the difference between NDM, OXA-48 or any of the other carbapenemase, again, to which we already saw targeted therapy should be given.

There are several tests now that are readily available for clinical micro labs to detect different carbapenemases. And so this can be cheap, easy and quick, multiplex PCR tests that are available and ELISA for carbapenemase testing. So again, giving the clinician the name of that carbapenemase is really, [00:39:00] really important in how they decide therapy. So if you don't have it on your rapid diagnostic, which I would argue on a blood culture might be really, really helpful and get patients to quicker therapy, if you want to do it on the backend and do it once you have culture, I still think this is a useful test. Part of the reason for this is that we know, and just to finish this out, that in an in-vivo model, there's a hint that things like OXA-48 when present, the MIC [00:39:30] might alone might not predict success.

What's shown here is a Murine Thigh Model out of Joe Kuti and David Nicolau's lab, and it's showing that here's the control, so how well these isolates grow with no antibiotic given, and here's isolates that are OXA-48 positive at lower MICs and with an MIC of less than or equal to one to meropenem when meropenem is given here, here, [00:40:00] here, and you start to get less killing and less inhibition, but that's fine because this is where the breakpoint sits. But for meropenem-vaborbactam, where the breakpoint is much higher because of the prolonged infusion and because against a KPC this is actually sufficient for killing, you see here that we're having less killing in this area, but we don't call this resistant until we're way up here. And so again, for my neutropenic patient, I'm not sure I want to use meropenem-vaborbactam, that wouldn't be really [00:40:30] active against an OXA-48, but we would still call it susceptible.

And so again, that's helpful information to know. And again, we don't have clinical outcome data proving that this is important, but this Murine Thigh in-vivo model is concerning. So in conclusion, I think that carbapenemase testing is really, really important and it's because many of the new agents were specifically designed to work against these carbapenemases. The role in rapid blood culture [00:41:00] diagnostics is available now and can help tailor therapy early on, which we know results in better outcomes. Unfortunately, with the emergence of new carbapenemases, the changing epidemiology in the US is making it critical to understand the enzyme, and I still do want an MIC and phenotypic susceptibility, but I'm going to interpret that in the context of specific enzymes as we have clinical data emerging. With this, I'll finish my presentation and thank you so much for the opportunity.

[00:41:30] Thank you so much for having me here today to talk about diagnostic challenges in multi-drug resistance. So we all know that this is a huge problem for healthcare, and so I think it's a really important discussion to have today. So here's my disclosure slide. So let's dive into that a little bit deeper. And so we know that resistance detection is best done whenever we use a multi-pronged approach. And so when we think about susceptibility testing, we want to be looking at both phenotypic susceptibility testing [00:42:00] and mechanistic testing in which we can do either by phenotypic methods or genotypic methods. And so this publication here by Idelevich and Becker, I think they did a really great job of exploring the different types of susceptibility testing and summarizing that for us. So generally speaking, we look at universal phenotypic susceptibility testing and that's what's most widely available in the clinical laboratory.

And then we can also look at detection of selected resistance mechanisms. So if we start with phenotypic susceptibility [00:42:30] testing, we know that that's universally applicable. So we can do that independent of the source or the specimen type and generally speaking the organism as well. So it's universally applicable to most things we'll see in the clinical laboratory. It's mechanism independent, so it doesn't matter the mechanism, we're able to take that organism and put it in presence of that antimicrobial agent and we're able to give a phenotypic categorization. So we're able to say, is that organism susceptible? Is it susceptible dose dependent, [00:43:00] intermediate or resistant? And that gives a great deal of information to the provider in terms of choosing the therapeutic relevance for that particular agent. But on the con-side of this, we know that culture is really slow, so it's dependent on microbial growth, which takes a long time.

And we also know that being able to detect resistance, it may require expression of a particular resistance mechanism. And so really sometimes this [00:43:30] can be expression dependent. So if we move to detection of resistance mechanisms, so here we're talking about maybe PCR or other types of tests that we look at in the laboratory, these are generally very rapid by comparison, so they may take 15 minutes to an hour, whereas with phenotypic susceptibility testing, we're generally talking about testing that's done within the timeframe of about eight hours to 24 hours. But again, another positive here is that we can actually use this [00:44:00] to confirm resistance mechanisms. So we may detect that we suspect that there's resistance there and we can use these genotypic or enzymatic tests, either molecular or enzymatic test to confirm that resistance mechanism. But there are a lot of cons for this too.

A negative doesn't necessarily always mean a negative. We can maybe get negative for a particular resistance mechanism, but phenotypically that agent may not work. You can also have positives that don't really mean they're positive. So here whenever we detect something [00:44:30] that is a resistance marker, it doesn't always mean that that organism is going to be phenotypically resistant to that particular agent. And really this is limited. We only have a few number of drug bug combinations in which we're able to test with these types of methods, but why do we still want to do it? So I listed a number of cons here whenever we talk about mechanistic testing, but there's really a number of reasons why we want to take this multi-pronged approach and perform both phenotypic and mechanistic testing. And so really [00:45:00] knowing these mechanisms supports infection control measures, so we're able to get that information out quickly and sequester those patients that really need to be on isolation.

Additionally, there may be treatment differences based on the information that we're able to provide on these particular mechanisms. So there's really a number of reasons why we need to explore doing both of these types of testing. So let's move on to some challenges that we're seeing. Our largest challenge is most definitely time to detection of resistance. [00:45:30] And so whenever we talk about culture, I already discussed that it takes a long time for us to be able to perform those, and so we can have pre-analytical delays, and that would really occur based on the collection and transport of the specimen. These aren't instantaneous. And then whenever that culture or whenever that specimen gets to the laboratory, excuse me, we know that it takes a little bit of time for us to be able to set those up into culture or perform a particular test. And then if we move to analytical time, culture's just inherently slow as we discussed.[00:46:00] But phenotypic susceptibility testing is absolutely crucial. And so whenever we talk about how long this really takes, it requires growth on solid media most largely, and that can take about 16 hours. And then we have to go on after we have that growth on solid media. So shown here is an example where you can see a colony on a plate where can go on to perform identification testing and then susceptibility testing. And so these can be done either consecutively or concurrently, but for [00:46:30] us to be able to perform that susceptibility test, this generally is going to take about 16 to 24 hours once we have growth on solid media. And so best case scenario for most culture types, we're looking at around two days before we can provide you that phenotypic susceptibility testing. And so the average turnaround time is actually going to increase when we encounter something that's resistant and that can maybe be three to four days or even longer, and especially that can happen if we have to send that testing out to an alternative laboratory.

So just keep in mind that this is really [00:47:00] comparatively slow. Then you can also have these post analytical delays. So even after we provide this information, who's going to be reacting to that result? Are they going to be seeing that? Are they going to be performing a change in the treatment of that particular patient based on the information that we're providing? So let's dive a little bit more into this analytical time and why that matters. So I just said that it's going to take us roughly two days before we can provide that information, and it can get even longer whenever we think that that there's resistance that's present. [00:47:30] So here's a publication by Bonine, in which they looked at, they specifically wanted to know whether a delay in appropriate antimicrobial therapy, which they defined as greater than or equal to two days was associated with worse outcomes.

And so this was really important because a lot of the publications have focused on blood cultures, but here they really looked at a high number of specific gram-negative infections in patients. So they evaluated greater than 56,000 gram- [00:48:00] negative infections. And what they found was a really a significant incidence of delayed appropriate therapy in these patients. So all taken, around 34% of patients had a delay in what they considered to be appropriate therapy, and that increased to around 46% of patients whenever they encountered resistant bacteria. And so you can see how that would be problematic whenever we're asking providers to make good empiric selections, and it really takes us a long time, it takes us these two days to be able [00:48:30] to provide that phenotypic susceptibility information for them to alter therapy. So we're really at a disadvantage in these patients. So the delay in this publication they showed led to poorer outcomes, and that included a longer duration of therapy, which is not surprising, right?

Because if you're giving your patient a therapy in which you determine to be not probably the most successful therapy, then you need to complete that course using the appropriate drug. A longer length of stay, so these patients ended up having to stay in the hospital [00:49:00] longer and it increased their likelihood of death or disposition to hospice care. So again, all negative outcomes associated with a delay to appropriate therapy. But this is really a balancing act because we want to make sure we are limiting our misuse and overuse of broad antimicrobial agents while still providing the best therapy to that particular patient. And so there are some cases in which those patients may actually need broad therapy, and we want to minimize our chance, [00:49:30] as we just talked about, of having worse outcomes by providing ineffective therapy. And so there's really this balancing act that we see.

So how do we go about doing that? Because we know that if we, based on the publication I just showed, that we can have longer length of stays and increase in morbidity and mortality and an increase of cost if we choose inappropriately. However, we know that whenever we use too broad of agents, we can increase our chance of hospital acquired infections. We have emergence of resistance. [00:50:00] We really lose our ability to lose those last line agents when resistance emerges, and it can add costs on that end as well. So how do we go about really balancing these things? And that's a particular challenge for us in medicine currently. So the answer here and which is a big area of exploration is being able to provide earlier susceptibility results to guide definitive therapy. So again, this is a large challenge for us, and one of the things that we can do as we wait for advances [00:50:30] in these types of testing modalities is knowing how to test accurately with the tools that you have in your laboratory.

I want to just take a minute to focus here on laboratories that utilize commercial AST systems and really highlight the fact that we have challenges with the technologies that we currently use. And so we may use just a single panel configuration. So my laboratory is an example of this, that independent of what source it is or what specimen type, what [00:51:00] is going on with that particular patient, we just have one gram-negative panel that we're going to run to look for susceptibility testing. So we're currently not able to optimize the agents on that panel specific to that particular patient perhaps. So we would know that if we were going to have a patient that was suspected of having a multi-drug resistant organism, then we would likely want to test some of these different agents. And so we're really limited in our ability to do that.

And then we also [00:51:30] know that between 38 to 70% of US laboratories, they're using obsolete or outdated breakpoints whenever they perform their susceptibility testing. And so what this means is that a patient may go to hospital A and be told that their organism is susceptible and then go to hospital B, and they're told that that organism with the exact same results is resistant to a particular agent. And so really there's a movement underway to get all laboratories using the same breakpoints and get everybody [00:52:00] up to date as much as possible. And so there's really a big push both from the commercial AST systems as well as governing bodies and all those in the clinical laboratory to really get everyone accurate with their susceptibility testing. So that's a great thing and a great thing that's happening currently. So as we wait for our current technologies to advance, how can we stay current?

We want to update our panels regularly to test newer agents and dilutions and use current breakpoints when reporting our susceptibility information. [00:52:30] If we're unable to perform the testing based on what's available in our laboratory, we want to define algorithms with our partners that are involved in this process. So maybe our ID physicians, our pharmacists, to rapidly perform reflex testing to get the right drugs tested so that we can get that information back to the provider. We want to explore and implement newer technologies for rapid detection as they become available. And so if you're not performing diagnostic tests directly from blood cultures, you really should look at doing that. [00:53:00] In addition, you want to be looking at methods to detect carbapenemases. So those are two ways in which those are current technologies that if you're not employing you really should look at doing in your laboratory.

So if you want to speed up susceptibility testing, one of the things that's important as well is establish that diagnostic confidence. Really explore what your system is doing well and understand the limitations and adapt to those. You want to communicate early and often. So whenever you suspect that you have a resistant isolate, don't continue to repeat test that isolate over and over [00:53:30] again. You want to let somebody know that while you're sorting this out, you suspect there may be some resistance that's there. You want to develop algorithms and strategize. This really requires a multidisciplinary approach. I just put an illustration here where you want to be communicating with your MD partners, the pharmacist, your antimicrobial stewardship team, infection control, obviously your microbiology team. And then keep in mind that at the center of this is the patient, how do we provide the best care to that patient?

So [00:54:00] because I'm a microbiologist, I have to plug us here, but we're really the first line of defense, those in the laboratory, at recognizing there may be resistance. So again, we need to really provide that competency training for those in the laboratory to make sure they're up-to-date and recognizing these resistance mechanisms. So complications and ongoing challenges, the cost is obviously an issue and will continue to be an issue, and that really limits the number and types of agents that we can routinely test and who we can test those on. So we can't perform susceptibility [00:54:30] testing over and over and over again on the same patient. And so we have to be really selective with how we go about utilizing those susceptibility testing platforms that are available to us.

So duplicative, unnecessary and expensive add-on testing can really add costs to the patient and the organization. And we need to make sure that we're meeting our regulatory requirements and those can be costly as well as we tend to add on different systems or different panels. As everyone is well aware, we have some workforce challenges. [00:55:00] And so that really is compounded by the fact that we need to provide additional training and susceptibility testing for our workforce.

And then the complexity, susceptibility testing is hard. So there's a lack of consensus as an example for defining multi-drug resistance, and that really exacerbates our difficulties in reporting and tracking the true number of multi-drug resistant organisms. We need to design and maintain algorithms. And so that takes some thought of how to report the current agents per source, per circumstance, [00:55:30] and with the correct interpretations, and we need greater access to training and competency materials. So those are some challenges that hopefully we're going about tackling in the near future.

So in conclusion, accurate susceptibility testing and reporting is just paramount to combating antimicrobial resistance. So if you're going to get something right in the laboratory, get this right. The greatest challenges for diagnostic testing include the time required for phenotypic susceptibility testing to get that result, [00:56:00] accessibility, the complexity. Again, these are very detailed tests, and so the complexity of that is a challenge, cost, and then again, workforce shortages. So in conclusion, I hope I leave you with this thought, the time to make your diagnostic improvements is now. So get current, stay current, and do what's best for your patient.

And I'll just add to that, our experience [00:59:00] over the last three, four years has sort of been just a case example of that where we've had primarily KPC driven resistance for our CRE isolates, but then because of the fact that we've had some of these novel agents on our panels or we've been able to set up disc testing for them early on when we need them, when we have carbapenem resistance, we're able to identify phenotypic resistance patterns that were really surprising to us where we saw resistance to some of these novel agents, which obviously that [00:59:30] is of grave concern. And thankfully, we've had great relationships with our clinical microbiology laboratory and Dr. Kern and the state laboratory, and then sent out these very problematic, scary isolates for sequencing and understanding what the mechanism is.

And that's where we're able to see that we're starting to see NDMs and OXA-48 and things that we haven't seen before and weren't ready to just expect with patients that didn't have major risk factors that we typically [01:00:00] would think of. So for us, exactly what Dr. Abbott was saying with phenotypic susceptibility, reliable phenotypic susceptibility for novel agents, and then being able to partner with folks to be able to send out for sequencing to understand what's happening here on just a broad basis, because our genotypic tests, some of our molecular tests are very specific and targeted, so we're not going to know what we don't know until we tap into that and figure out what's going on from a patient to patient standpoint [01:00:30] or even a surveillance standpoint.

Yeah, so it's a great question. We have molecular panels directly from blood because again, with that particular source, you're often getting a fairly pure culture, so you kind of know what you're working with. It also has the ability to amplify that organism in that blood culture bottle. So it's kind of growing it up in advance. So we have that. And then in addition, [01:01:30] we have a molecular test that we perform where if an isolate meets certain resistance patterns, we will set that test up. And so we know that pseudomonas aeruginosa or Acinetobacter often will have other resistance mechanisms that are not carbapenemases that will cause them to be extra resistant.

And so we kind of limit the type of testing that we do based on the organism and the pattern that we see and then also what we already know about that patient. And so we have these built in [01:02:00] to where the phenotypic result comes across, let's say it's a urine isolate or a wound culture. And then we're able to say, okay, now this is an isolate that needs further testing for us. And so we would repeat that susceptibility, but also set up some of this additional testing at that same time so that we can confirm the phenotypic susceptibility result plus get a mechanistic result. And so that's how our laboratory goes about doing that.