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]
Emily: 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. I am Emily McDonald, general Internist and Infectious Disease Trialist, coming to you from McGill University in Montreal, Canada.
And I'm an associate editor at CMI Communications. Today's episode is the first in a recurring series called The Quarterly Catchup. So here's what you can expect for quarterly catchup episodes every three to four months, we'll have CMI comms editors. Pick one to two articles each that have been published in the past three months.
We will give you our take on the studies and catch you up. we know that everyone has busy schedules and it's hard to keep track of all the science being published. We'll aim to keep these episodes short around 45 minutes so you can hop on the treadmill or the train or get ready for your commute.
We've got you covered [00:01:00] with a rapid fire update on the latest research. Unlike some of our other summary episodes such as the annual top 10 or our post ESCMID episodes, we might not always focus on clinical trials. Depending on which editors are hosting. We'll pick topics that we find personally interesting, so you can look forward to some variety in the science we'll be presenting on the quarterly catchup series. Today my co-hosts are my awesome fellow editors at CMI Comms, Thomas Tängden and Nav Narayanan.
Thomas: Hi everyone. I'm Thomas Tängden, an infectious disease specialist and a professor at Uppsala University in Sweden. I'm interested in learning more about how we can use antibiotics better to improve patient outcomes and prevent adverse events. Therefore, the studies I picked for the today's episode is on collateral damage on the gut microbiome following intake of oral antibiotics and a preclinical study on combination therapy for multi-drug resistant gram negatives.
Nav: Hi everyone. It's nice to be here with Thomas and Emily. I'm an infectious pharmacist at Rutgers University and also an [00:02:00] associate professor there at the School of Pharmacy. I'm interested in tuberculosis as well as HIV and drug resistant bacterial infections.
so for today I picked a trial, which is the first genotype stratified randomized control trial for TB meningitis and then another trial. That was published, very recently in New England Journal of Medicine, looking at long-acting injectables for, HIV.
Okay, great.
. I'll go first with the two studies I picked. The first study I picked, is because of my hatred of mosquitoes and also because I saw a lot of opportunities for mosquito puns. So, this was dengue suppression by male, Wolbachia infected mosquitoes, published in the New England Journal of Medicine in February.
Emily: By first author Jue Tao Lim and Colleagues. Dengue is becoming more common because of climate change and urbanization, and there are very limited options to prevent or treat it. There have been trials of repellents that are somewhat effective, but chemicals have unwanted side effects, [00:03:00] so very cool. We have learned that Aedes aegypti mosquitoes that have been infected with Wolbachia pipientis bacteria may reduce transmission of dengue virus.
This is an approach that's been tested, whereby they released male and female infected mosquitoes to replace the wild type mosquitoes. So this worked in a previous cluster randomized trial in a city in Indonesia, but apparently the protective effects can vary. Okay, so these scientists were itching to try out a different technique.
The approach they studied was to irradiate male A. aegypti mosquitoes that were infected with Wolbachia bacteria and release the males. Apparently this can be more effective, than the other technique. And this is because when you release infected lady and male, mosquitoes, there can be improper sex sorting, and the mosquitoes can still mate and become established in the region.
But if you irradiate the infected male mosquitoes and release only infected males, they can't [00:04:00] have viable offspring with uninfected females. And if you do this repeatedly, so you release them over several cycles, it's possible that it could take a bite out of dengue transmission. As these authors put it so eloquently, this trial builds on baseline entomologic and epidemiologic field studies bugs infected with bugs to kill viruses.
Genius. This was a cluster randomized trial that took place in Singapore. There were 15 clusters and eight clusters received infected male mosquitoes, and seven clusters received none. The primary endpoint was symptomatic dengue, and what they found was that the percentage of residents in the intervention clusters who were dengue positive was lower than the percentage of residents in the control clusters.
So out of 5,722 tests, there were 354 cases of dengue. So that's about 6%, in the intervention arm. And then in the control arm there were 7,080 tests [00:05:00] performed of which 1,519 were positive. So that's 21%, so a 6% rate of infection in the intervention versus 21%, rate of infection in the control. So the protective efficacy of the intervention actually ranged from about 71 to 72%, with three to 12 months or more of Wolbachia mosquito exposure with an odds ratio of 0.28, to 0.29.
So seems to be pretty effective. my question to my co-hosts, should we invest in designer mosquitoes?
Nav: I'm just gonna say that your puns are on point. I usually don't give you credit for your puns,
Emily: Thanks Nav,
Nav: but they were on point for this one. I found this so fascinating though when I've seen this in previous trials. how you get to this point of thinking about infecting mosquitoes to prevent a transmission of a different pathogen is just so genius to me in that way.
And just shows how [00:06:00] much you need to invest in sort of innovations. 'cause you never know when a brilliant outcome is gonna come out from this very innovative thing. So, I, love this study. I'm, glad that you presented this.
Emily: I totally agree. I remember when I first read about this technique, I thought, this is the type of innovation that we need in science.
very cool. And I love that they're continuing to still think about what's the best way to optimize it now.
Thomas: Yeah. And to me, I also love this study, and I think it's interesting because it brings sort of a new approach or a new concept to preventing disease. there are two components that are interesting to me.
first, the thing that we're using microorganisms to, treat microorganisms like we would do in bacteriophages, but this is also about prevention. How can we actually prevent disease rather than treating? it's cool.
Emily: I also like that they use cluster randomized, in order to deploy the intervention.
Kind of cool to think about, you know, how you're gonna actually get the mosquitoes out there and how you can have a control. Okay, I'll move on to my second study, which [00:07:00] was, a re-analysis of two trials. So it was adjunctive phosphomycin for the treatment of Staph Aureus bacteremia, a pooled post hoc analysis of individual participant data from two randomized trials.
This was published in CID by first author Francesc Escrihuela-Vida, and colleagues Staph aureus Bacteremia has high mortality. Many combination therapies have been tested without success. So we have RCTs of rif, gent, ceftaroline based therapies, vanco, dapto, all have failed to show benefit and several have actually shown increased harm.
Phosphomycin blocks peptidoglycan synthesis and has some synergistic bactericidal features and anti biofilm properties. So another in the long line of combination therapies to be tested for staph aureus bacteremia. There were two fairly recent RCTs that tested phospho combination therapy. So I'll tell you about the individual studies and then we'll talk [00:08:00] about the combined analysis from the individual patient data.
So BACSARM was the first study. It compared high dose daptomycin, so that was the backbone therapy. And by high dose they gave 10 milligrams per kilogram per day in combination with phosphomycin, and they compare that against daptomycin monotherapy. And this was for patients with MRSA bacteremia. The second trial was SAFO and that evaluated cloxacillin plus Phosphomycin versus cloxacillin alone.
For MSSA bacteremia, neither found a benefit for treatment success or mortality. Both seemed to suggest that it could decrease the duration of bacteremia, and both studies were stopped early. So the present study pooled individual patient data from both studies. The authors combined the intention to treat populations.
So these were MRSA and MSSA bacteremia patients. They got at least one day of the study drug. It included complicated and uncomplicated bacteremia. Both studies excluded patients with endocarditis. [00:09:00] BACSARM obviously also excluded MRSA pneumonia, because daptomycin can't be used to treat MRSA pneumonia.
And they gave different amounts of phosphomycin and for different durations. So that's fairly important to be aware of. BACSARM gave phospho at two grams every six hours, infused over 60 minutes, and patients received the combination phosphomycin for the whole study duration, so longer durations of, phosphomycin.
SAFO gave phosphomycin three grams every six hours as a four hour infusion. And, in case you're interested, the cloxacillin was at a fairly high dose, so two grams every four hours, and they gave the combination, part of the treatment for seven days. And actually in the second study in SAFO, clinicians were actually instructed, that they should watch the volume status and the salt load, and that they should give potassium and that they should give Lasix as needed.
Whereas in the BACSARM trial, there wasn't any such instruction. [00:10:00] So the authors of this study in CID combined the individual patient data and they looked at a primary outcome of treatment success at eight weeks. So this was a composite outcome of alive, no signs of relapse and improvement in clinical signs and symptoms.
They also looked at, several secondary outcomes. So for example, all cause mortality. They looked at this at different time points, and they also looked at adverse events, and they took a Bayesian approach to their combined analysis for the primary outcome. They also performed a frequentist analysis. for the sake of time, I'm gonna focus on the Beyesian analysis.
And as a reminder, this approach for some people can be more intuitive, when interpreting whether the probability of a certain treatment is superior and it relies less on a rejection of the null hypothesis or what some people might consider to be arbitrary p value. They used a minimally informative prior, so fairly neutral.
And then they also [00:11:00] carried out a sensitivity analysis among others. So several sensitivity analysis, but the one I'm gonna highlight was that they also redid the analysis where they used a prior that was pessimistic. So that was, assuming that phosphomycin was harmful, what would the results look like?
So the combined analysis included 369 patients. 178 received phosphoymcin combination therapy and 191 received the backbone monotherapy. the groups were fairly similar in their comorbidities and the primary analysis found a 91.8% posterior probability that combination therapy improved treatment success at eight weeks compared with monotherapy.
The pessimistic prior, remember that's where they made some assumptions that phosphomycin might actually be harmful, yield a reduced probability, but still of superiority, so it reduced the probability 75.8%. subgroup analysis actually suggested that there might be a greater probability of benefit for [00:12:00] patients with MRSA bacteremia.
Emily: so for the MRSA subgroup, the probability of superiority was 92.6%. And for MSSA subgroup, it was only 73.7%. and nosocomial acquisition also looked like there was a higher probability of superiority from the combination therapy. So when they looked at the nosocomial subgroup, there was a 93.8% probability of superiority versus, actually quite a reduced probability.
So only 47.9% in community acquired and then finally, the relative risk of adverse events leading to treatment discontinuation was clearly higher in patients with combination therapy with phosphomycin. So the relative risk was increased to 1.97 with a credible interval of 1.04 to 3.92.
And this was particularly true in the BACSARM trial. So the posterior probability of harm was 99.3%, compared to the SAFO trial where the [00:13:00] probability of harm was 72.1%. And just, you know, as a reminder, those two studies, gave different doses and durations of phosphomycin and the SAFO trial where the probability of harm was, slightly less from this analysis.
That's the one where they gave the phosphomycin for a maximum of seven days, and the clinicians were also instructed to watch the volume status and to give Lasix as needed. Okay, so I'll stop there. And I am interested to hear, nav and Thomas's your impressions of this study.
So this is interesting. We've tried this in my hospital a few times to add on phosphomycin in some difficult to treat patients with staph aureus bacteremia or endocarditis. And we've had some problems with side effects. So I can relate to the discontinuation of treatment We've done that because of heart failure.
Thomas: Like it's a lot of volume and also potassium challenges sometimes in these patients. But it's an interesting drug and it's certainly [00:14:00] has some interesting features also in biofilm infections, for example. when you told me about the results, I thought that MRSA and MSSA patients might. Slightly differ, in terms of infections or comorbidities.
And also I assume that MRSA is more commonly encountered in healthcare associated infections than community acquired infections. So how much do you think that differences results, for MRSA versus MSSA could actually be also about the type of infection, maybe virulence factors of the bacteria or the fact that they're the second drug in the combinations is different.
You have a beta lactam versus glyco peptide type antibiotic, do you think?
Yeah, so I have no clinical experience with IV phosphomycin, here in the, us. So every time I see these trials, I always wonder how much it's actually Implemented clinically. And so just to even hear Thomas, your, experience helps paint the picture for me a little [00:15:00] bit as to how, it gets used.
I don't think there's a lot of questions of, are some of the agents that we have for staph aureus bacteremia, effective. But the problem is that in my opinion, it's just not effective enough. Right? There's still so much morbidity and mortality related to staph aureus bacteremia, that there's always some, room to improve in terms of how efficacious our agents can be.
Nav: And so it's nice to see these, trials, and I actually enjoy seeing them being analyzed in this way. Like Emily said, I, think we need to start thinking a little bit more innovatively, not just about The technology or the therapy, but also how we analyze these trials. And so I've learned a lot from Emily and, just how these approaches are, more intuitive way of looking at it.
And I think it contextualizes, how we apply this in clinical practice. So from a methodology standpoint, I really find it fascinating and, pretty impressive as well. I guess I want to ask you both. What do you think about, these [00:16:00] drops in the posterior probability from going above 90% to in the seventies?
I mean. Is there a place to draw the line, or do you just kind of think about it as a continuous variable? But how do you start interpreting it as things fall?
Like, is there not a cutoff, but just, you know?
Emily: we're doing it intuitively at the bedside when we have access to different treatments. Uh, sometimes we're doing it without any data, right? So we're saying, okay, well we know that, phosphomycin has a good effect on biofilm, so I'm gonna try and use it in this really sick patient and, see, see if that makes a difference.
Um, so so Now you know that it's probably better, but than, giving it alone. You know, if you believe the study, maybe for MRSA patients, if you also, treat the incurring heart failure. but that, that probability is maybe somewhere in the seventies for MSSA, or maybe it's somewhere in the nineties for MRSA.
So [00:17:00] it's more information than you had before, but you still have to make a clinical decision. I think people have started to talk a lot about, where is that cutoff? Actually, other people are interested in whether or not there's a cutoff for the probability. we want to dichotomize because that's how we've been trained to interpret research.
some people have talked about if you're bringing a new drug to market, maybe that probability needs to be set higher. Maybe you need to have more than a 95 or more than a 99% probability, that the drug is non-inferior or that it's superior if it's not inferior with a certain. non-inferiority margin.
But if you're comparing two standard of care therapies that we're already using, maybe it's enough to say that there's a 75% probability that this is superior since we're using both anyways. And then also maybe it matters of course what the harms are. so, you know, let's say it's 75% probable to be superior if you use the combination therapy, but 99% probable [00:18:00] to be harmful.
And in the end, at 30 days, that shakes out to, increased mortality from harm. Like that also matters, right? And so you would also need to follow up on the all cause mortality because if you are 99% probable to cause heart failure, But you can reduce clinical success at eight weeks, but then people die by day 90.
that's relevant too. So I think you have to consider the harm you have to consider, whether it's a new drug being brought to market or whether you're doing head-to-head comparative effectiveness. Kind of a long answer, but just my thoughts on those things.
Nav: No, that's helpful, at least for me, just to hear how, you think about it as well.
and it makes me appreciate these initial priors going in and the assumptions of is it neutral? Is there some benefit based on prior data, or is it, like you said, a, a pessimistic prior? Were there some concern for harm as Thomas even pointed out in, his clinical experience that, there's some tough things to deal with when you're giving that much phosphomycin, the electrolyte imbalances.
And the [00:19:00] reason I ask is kind of what you alluded to is just, eventually when these things catch up from like a regulatory perspective and bringing new drugs to market, and if this is a more intuitive, and a more, valid way that speaks to what's happening clinically, at some point, regulatory agencies are gonna have to put out guidance or think about these types of things.
So I just wondering, if there's a buzz in sort of the clinical world, then hopefully that informs some of these regulatory, parameters that get put probably in the future sometime, if that happens. .
Yeah. And also there are two levels of uncertainties.
Thomas: So you have the probability of benefit, but you also have a composite endpoint, right? So is the benefit more towards mild adverse events or is it about mortality? So that could also make a difference, I guess, in the end for a clinical decision. But I agree. It's interesting approach.
Emily: Okay, well then let's move on to Thomas, who's gonna present his two articles to us now.
Thomas: Okay, moving on. My first article was just published in Nature [00:20:00] Medicine by first author Gabriel Baldanzi and his colleagues, and the title is Individual Level Prescription Data of 14,979. Individuals reveal links between antibiotic use in the past years and the Gut Microbiome composition.
So the background to this study is that any systemic antibiotic use will cause damage to the gut microbiome. And we know also that perturbations in the microbiome are changes in its composition, such as reduced diversity have in turn been associated with chronic health conditions such as diabetes or inflammatory bowel disease.
And in this study, uh, they combine data from existing population cohorts, including a total of almost 15,000 adults living in Sweden. And, all participants were asked to provide one fecal sample. And these samples were then analyzed by sequence based methods to determine the composition of the guide microbiome, like which species [00:21:00] or bacterial families were present and at both levels.
And then they compare this to expensed, antibiotic prescriptions in Swedish pharmacists to see if there were any associations between previous antibiotic use and the results of the microbiome analysis. The classified antibiotics into 11 classes, depending on the spectrum of activity and mechanism of action, and, tested against prescriptions or antibiotic intake in the last year, the last four years, or the last eight years, to see how much the microbiome had changed and if there were still traces of antibiotic therapy in these samples.
And what I think is interesting with the study is that you can see very long-term effects with some specific antibiotics. So, the strongest associations were found for Clindamycin and Quinolones, where you could see even after four to 10 years, that these individuals had received antibiotics. And, as an example of an output from the [00:22:00] study, each course of Clindamycin less than one year before fecal sampling.
Was associated with 47 fewer bacterial species detected. And each course of quinolone was associated with 20 fewer species detected. as expected, more narrow spectrum drugs such as penicillin, we had minor effects and not as long lasting. an unexpected finding was that oral Flucloxacillin, which is a narrow spectrum drug, had a quite pronounced effect, almost as bad as Clindamycin Oral clinical.
And , it's a finding that's difficult to explain, but could be related to the pharmacokinetics of.
flucloxacillin resulting in very high drug concentrations in So these are the main findings of the studies. Emily and Nav, do you think these kind of studies are informative and should guide clinical practice or any other thoughts
Nav: My personal opinion is, and I'm biased because I very much am interested in, the [00:23:00] microbiome as well as, the effects of the microbiome on the heterogeneity of treatment response and some area that I try to focus on for research. So do I think that these.
Results are important. I do. It's, very hard for it to directly impact clinical practice at this level. Do I think it's something that should be in the back of, clinicians' minds in general? I do. Just the same way that, you should be thinking about environmental sustainability in terms of IV to PO amongst those clinical decisions.
So there's a lot of things that I feel like should be, seeded in the broader thought of how you make decisions. But is it gonna necessarily make a very specific change in, patient to patient? Maybe not so much at this level, but I think the more that we learn about this, it sets a new scientific context of how we think about the benefits and harms.
At a broader level, right, in terms of antibiotics and, drug choice and whatnot. So, [00:24:00] maybe not at a patient level, but I do think these types of studies are, really important. because that also, affects, how we do microbiome studies in general, right? Because of how long lasting some of these effects are.
Because we usually think, six months out the microbiome sort of comes back to its baseline or whatnot and in some of these studies, but maybe that's not really the case. And so it, it's an important sort of consideration I think when you're doing microbiome studies as well and thinking about how that leads to other disease processes.
Thomas: Yeah, like you say, there were no clinical outcomes in the studies, just. observing changes in the presence of bacteria species, many of whom we can't cultivate, or many of whom we don't really know the biological function of. So there's definitely a gap from this type of observations to clinical importance that we still have to learn more about, I guess.
Yeah, I think we're really early on in what we know about the impact on the microbiome. I'll add two things. One, I wonder [00:25:00] what patients would think of this data if they knew that taking an antibiotic could affect their microbiome for four to six to eight years. Could be something really important for the public to be aware of often.
Emily: People go to the doctor kind of maybe requesting antibiotics in the outpatient setting. So something to be considered about how to translate this information out to the public. And then I'll make one incendiary comment, which is, you know, I'm a big proponent for early oral switch, but maybe we need to learn more about the difference between oral versus intravenous antibiotics on the gut microbiome.
Just, you know, something for us to keep in mind for future studies.
Thomas: Yeah, those are two very important points and I think we learn from a national initiative on communication of, the risk of taking antibiotics to patients. That this is a very strong message. If you tell a patient that we don't think that antibiotics will help you, and you can probably have traces of this for years.
Like especially if if you give them a quinolone as a treatment option, many of the patients [00:26:00] with actually choose not to take antibiotics in that situation. Even though we don't know exactly what the implications are, but just the fact that it could be potentially important. and the second point is also something we're discussing a lot in stewardship, discussions here in my institution, like you said, early Oral switch has always been one of our main messages.
But actually, if you think about the gut microbiome, oral drugs are likely to give higher drug concentrations in the faecal, setting. So, for sure it could be the opposite, in terms of selection of resistance that we anticipated. So it's difficult to find out more about this.
Nav: Yeah, I, find it pretty interesting, how there's.
Continued competing interests, right? In, in the benefits of one treatment strategy and then maybe some signals or thoughts that maybe there are not necessarily harms, but not, it's just a hundred percent beneficial, right? Like if you have someone [00:27:00] that is at very high, you know, they're 85 years old, they weigh 45 kilos and are you gonna switch 'em to quinolone for three more weeks of therapy?
And you're worried about adverse reactions, it's just at the individual patient level, you, have to start weighing some risks and benefits. And this just seems to add something in general. I mean, we're very far from having a very concrete, you know, harm understood of these long-term effects, from, from this particular study.
'cause like you said, Thomas, obviously it's not looking at, clinical outcomes, but the point that you made about patients and Emily, that you especially made about patients and. Them understand the risk of antibiotics. I actually think this is pretty helpful in that way because I see more and more microbiome in the mainstream communications and, media.
And so I think patients and the general public and, you know, the general communities that we serve are becoming more and more aware of, and maybe they don't necessarily know the scientific details of how [00:28:00] disturbance in the microbiome could lead to chronic diseases, especially early in life and all these other aspects.
But they hear the microbiome. I mean, I see commercials of like soap and body wash talking about the skin microbiome, right? And, and so, there's whole, all those drinks that they show with, you know, probiotics and it's improving the health of your microbiomes. I think it's mainstream now. So even though there's other maybe more clinically definitive reasons why to avoid antibiotics, this might be a way to really.
Open the eyes of patients to say like, you know, you have to think twice before you take antibiotics if there's not a really strong indication for it. So, you know, it might have some other patient communication benefits. I think of this type of work.
Thomas: Okay, let's move on. a second study is a preclinical study on antibiotic combination therapy from Quentin Vallé and his colleagues published in CMI recently, and I apologize beforehand, this is too technical for some of our listeners. The title is [00:29:00] evaluating the antibacterial activity of Ceftazidime/avibactam and aztreonam Combinations against Multi-Drug Resistance Stenotrophomonas maltophilia complex isolates in a hollow fibre infection model.
So, the background to this study is obviously merging multi-drug resistant gram-negative bacteria on the lack of new antibiotics. even with the newer drugs that we got the past 10 years or so, we sometimes find ourselves in situations where there are actually no available drugs that seem good enough for monotherapy, and that's when we move to combination treatments.
In this study, they targeted stenotrophomonas maltophilia, which is intrinsically always resistant to carbapenems and usually also resistant to many other treatment options such as amino Especially in patients where cotrimoxazole cannot be used due to resistance or intolerance, combination the therapies are probably needed to, achieve additive or [00:30:00] synergistic effects.
Thomas: So what they did was the in evaluated combination of aztreonam together with a fixed combination of ceftazidime/avibactam in vitro. It's a preclinical study and they compared it to the standard treatment of Cotrimoxazole. they used 11 strain sample from patients in the US and first did more simple timescale experiments with static meaning constant concentrations throughout the experiments.
Then they created a PK/PD model mathematically based on these data and made predictions of. The synergistic effects of the combinations or whether they're likely to result in bactericidal effects or synergy at concentrations achieved in treated patients in blood. And finally, they made also in vitro experiments, using two of the strains in a hollow fiber infection model where you could add dynamic drug concentrations that may make exactly the kind of pharmacokinetic profiles that you will [00:31:00] get in patients.
outcomes in this study is described as reductions of the area under the log, transform bacterial count and emergence of resistant bacterial population. So this means it's a measure of the amount of surviving bacteria throughout the experiments and also. If you're likely to get, resistant populations of bacteria during therapy.
the results were promising for the three drug combination of aztreonam and ceftazidime/azobactam when you compared it to cotrimoxazole. You could see a better bacterial killing after 24 hours, and also suppression of the most resistant emerging populations. in the end, the authors concluded that this combination seems to be promising and should be further tested in animal models, and ultimately they suggest we move towards a more individualized approach to treatment where you can take into consideration. Patient factors, therapeutic drug monitoring, but also what kind of resistance mechanisms the particular [00:32:00] pathogen has.
Thomas: So this is far away still, but it's an interesting approach. This is close to some of the research I do myself. So I found this interesting. what do you think about this
Emily: I have had patients where I had no more options. within the past year I had a patient with Stenotrophomonas, and there was nothing within Canada that we could treat them with. So I think these studies are important. And again, when you're left with no options, on social media, I see people in the infectious diseases community posting the resistance profiles of bacteria saying, what should I do?
so we need studies like this. It's more information than no information at all. It has to start somewhere. So I think it is relevant.
Nav: Yeah, we see this clinically, and even though this isn't a randomized clinical trial and, you know, uh, Emily redid a Bayesian analysis of it, this is still a useful.
Study in the sense that it gives you information on understanding what, at least the [00:33:00] activity is, right? Because intrinsically you want to know that something is active and, and how you describe it. And, and I know this group does really good work in that, in that space, uh, in the sense of understanding the mechanisms and developing the PKPD models to help, characterize that activity through those mechanisms.
So I think it's important that these types of studies are done and, and then they better inform, in vivo models, clinical studies, and it's the right way to do things, right, when you're trying to develop a scientific program to help understand this stuff and optimize itself. Thomas to you, what is most exciting when you see this type of work?
Thomas: Hmm. I think. What's important here is that in vitro you can play around a little, you can try different types of pathogens. You can stratify them by which enzyme they produce, or if they have efflux pumps that are active or decrease permeability changes.
So you can try to figure out why does this combination work against these bacteria that seem to be similar, but they're not [00:34:00] genetically. and you can try out different dosing regimens and Importantly, you're free to combine whatever drug you want to with whatever other drug you want to combine it with.
And also enzyme inhibitors you can play around with a little bit. we're relying a lot on beta lactam, beta-lactam as inhibitor combinations now for multi-drug resistant gram negatives. But they always come in a fixed combination, which might not be perfect for that particular pathogen you're treating.
Today. this is one example where they actually used one of the fixed combinations, ceftazidime avibactam, and combine it with a second beta lactam. So it's a Double beta lactam therapy. some people might argue that you don't need ceftazidime. In this case, the authors kept it because they think it might add synergies to the combination, but maybe it would've been even better to combine it with meropenem, avibactam and aztreonam.
Who knows? So I think this is a, very odd topic, and it's something that, several [00:35:00] research groups w right now to see, can we use the newer drugs in combination with other enzyme inhibitors or old drugs to improve their efficacy and also importantly to suppress emergence of resistance because I think, many people.
Are a little bit disappointed with the new drugs. They're not as effective as we had thought. And we also see quite a lot of resistance development already. So we probably need to apply combination therapy with them also in some situations. So I think this is an example of type of study we'll see more of.
it's important to point out of course, that the situation in vitro is very, very different from a patient. So there are several steps more to take. you have a lack of immune response, for example. So I think we probably overestimate the risk of resistance development in hollow fibre experiments that are very long in their duration.
But, it's the first step. [00:36:00] It's interesting.
Okay, we'll move on to navs, two studies now.
Thanks. The first study I'm gonna present is, a study that was published in New England Journal of Medicine, February 18th, 2026. Um, so this is by Rana and colleagues from the la This is the Latitude Trial Team. Uh, the title of the article is Cabotegravir Plus Rilpivirine for Persons with HIV and Adherence Challenges.
so just some context, especially those that are not as well-versed in HIV. While antiretroviral therapy has been revolutionary, at least in the US there's still a large portion of people with HIV that don't have viral suppression. So the estimated number of people with viral suppression is about 67%.
Nav: and there's a lot of reasons for this. And one of the big ones is, adherence to their antiretroviral therapy because of structural barriers and adverse events. and there's other competing priorities that, reduce the adherence and then therefore the, success of viral [00:37:00] suppression.
a few years ago, the US FDA approved these long-acting injectables, and specifically what's being studied here is Cabotegravir plus Rilpivirine. but it was specifically approved for people with virologically suppressed HIV infection that was already receiving oral antiretroviral therapy. and so since then people have.
Thought about using this in other contexts because in the phase three trials, they excluded people with active viremia. but there were some data, some case series, some observational work that showed that you can actually get viral suppression with these long-acting injectables in these harder to treat populations.
And, and as a result of that, I think it was pretty, forward thinking. The treatment guidelines in the US recommended considering these long-acting injectables in patients, who have inadequate adherence to antiretroviral therapy. They have advanced disease, limited treatment options, but the evidence base for that was.
[00:38:00] Very little and scanned. And so it was kind of an expert opinion for these very difficult situations. So this trial is the first randomized control trial. it's a phase three prospective multicenter, open-label, randomized control trial. The goal was looking at safety and efficacy of these monthly injections, of long-acting cabotegravir and rilpivirine, with daily oral antiretroviral therapy.
And so they're comparing standard of care to this innovative long-acting injectable method, and specifically in these patients with adherence challenges. so these were adults that were included. They had no relevant resistance mutations and the way that they were defining non-adherence.
Patients with poor virologic response, with an active prescription for antiretroviral therapy. They had lost a clinical follow-up with non-adherence to their antiretroviral therapy or, both. And this was happening for a while. So at least six consecutive months, all within a year and a half of their enrollment in the trial.
So patients with [00:39:00] challenging, situations that led to non-adherence to their therapy. And the reason I picked this study as well as the next one we'll talk about is the design and the population they included, I think is pretty interesting. So right off the bat, you're doing a randomist control trial in a challenging clinical situation in patients with non-adherence.
and they've been on therapy for, quite some time. And the other thing is their trial design was, very interesting. So they had two steps. This first step is they enrolled all these patients. they essentially kind of got them on standard of care therapy, for some time.
And so they gave them oral interruptive vault therapy for up to 24 weeks. They gave them conditional economic incentives and standard of care adherence support and counseling. At that point, those that met virologic suppression were then randomized to the long-acting injectable or just continuing standard of care at that point.
If you're looking at the paper, the first figure I think Is the best to just [00:40:00] see the design very easily. So in step two, they randomized them one-to-one. They gave them the long-acting injectable with the oral lead-in or the continued standard of care, oral, antiretroviral therapy.
the punchline really is that these patients, had superior, success with the long-acting injectables. So the regimen failure that occurred happened significantly less in the long-acting cabotegravir rilpivirine arm. So it was 22, about 23% regimen failure versus 41% regimen failure in the standard of care arm.
So it was about an 18% difference between the two arms. So clearly there was superiority demonstrated there. And if you look at the Kaplan Meier curve, that superiority, the curves start to separate very early on. So around week four, week eight, the curves of regimen failure start to separate.
So clearly there's some benefit that starts to emerge relatively early on, and these are monthly injections, right? So already [00:41:00] after getting two injections, you already start to see some, benefit relative to just continuing oral, standard of care. I thought this was very interesting that they were able to apply this long-acting therapy finally in a population that you might see more relevantly clinically as opposed to the same old population of those that are virologically suppressed from oral therapy, and you're just switching them, right?
It seems like if they're biologically suppressed, you can switch 'em to so many different options and they'll do okay. But here if patients with real adherence challenges, Just to paint the picture , and then I wanna hear your guys' thoughts on this. at the baseline, HIV level was greater than 10,000 in about 41% of the population in step one in the screening population.
most of these patients were in a relatively older age group. There were, 26% of them were over 51 years of age. So it also speaks to sort of [00:42:00] demographic shifts in, in the aging population in the us. and then the median duration that there were all antiretroviral therapy before trial entry was seven years.
Nav: And the range, the intercourse trial range was three to 13 years. So these patients have been, you know, with their HIV diagnosis for, many years at this point, and still sort of struggling with adherence issues. And then the last thing I liked is they, they screened these patients for social determinants of health that would influence their.
biologic success on antiretroviral therapy. And I thought that was, something that's long overdue in these types of trials where clearly there's influence of social determinants on the outcome of these patients. So, this was an important trial for the design, the population, and then obviously the findings.
So curious to know what you guys think about this.
Emily: well a big part of my research is, reducing polypharmacy or taking multiple medications. So studies like this I always find interesting because patients, people with HIV are, a, population that take. Multiple [00:43:00] medications, often 10 medications a day. And so if it works for virologic suppression, it also can address, pill burden.
it might mean people are more inclined to take some of their other medications, maybe to treat comorbidities like diabetes and hypertension. If they're having to take, five pills a day instead of, eight or 10 pills a day. I think that that can make a difference. So I think it's, patient, oriented, in terms of the regimen.
And then, we hope that also these medications will be, uh, affordable, because, the patients that they're being tested in sometimes, because of different social determinants of health might not be able to access these medications. So I would also hope that that's, you know, addressed in the future.
Yeah, just to. speak to that point. in the table one, right , where they're describing the population, they had annual household income and 29% in step one in the overall group, 29% of them had an annual household income of less than $5,000. clearly that speaks directly to your point, that to implement [00:44:00] this, there can't be the mismatch between the innovation and the access from patients as well as the healthcare systems that either allow or don't allow access.
Nav: So obviously this was speaking to the, US and, our healthcare system. It speaks for itself. I'll just put it, that way in terms of, facilitating some of these things in terms of some patients with insurance, how things are changing at the state level in terms of what HIV medications are accessible and non-accessible.
and there's been some pretty dramatic changes in some states, recently, which, really is disappointing to see and speaks to that point. Emily,
Thomas: I'm not really experiencing treating these type of patients, so forgive me if my comments are basic. my thoughts were first, were there any other interventions made, like, except from switching to IV in terms of follow up or healthcare visits that could explain the difference. And my second question was [00:45:00] touching on what Emily just asked.
Apart from cost, are there other reasons to not. Implement this now that we have this study showing, uh, such a big difference or improvement outcomes.
Nav: Yeah, it's an interesting question. they talk about how some of these conditional economic incentives, that was really in like step one.
and so that was limited in, the trial. And so the financial incentives were not included in step two, which is when they randomized them to continue standard of care or go on to the injectable. So it's one thing that touches on that.
In terms of implementing this, one thing is any long-acting injectable, I, think it takes standing up in whatever clinic or whatever program it is, because of, getting access to the medications, the administration of the medications.
Nav: It's different than when you write a prescription for oral therapy and then the patient goes and picks it up, outpatient, the patient is taking it at home as [00:46:00] opposed to coming in, getting the injectable, getting the injection once a month. who's gonna do the injection, the ordering of the drug in the clinic.
So there's some logistical things that I think always need to be taken into account, but. Uh, you know, this has been shown at least proof of concept to be implemented in tough to treat situations. in some of those case series, it was patients that had housing instability and, they had active viremia still.
And so I, think there are some challenges and logistical barriers that have to be considered, but I certainly think it can be implemented if there's sort of effort put into to that. I don't think it's, prohibitive to be able to do this for patients in a very standard clinical setting.
All right. So with that, let's move to the second paper that I thought was quite interesting. So this was a paper published N it was published online January 15th, 2026. the title is Genotype [00:47:00] Stratified Adjunctive Dexamethasone for Tuberculous Meningitis, and HIV Negative Adults, A Randomized Controlled Phase three trial.
this was out of the Oxford University Clinical Research Unit in Ho Chi Minh City, Vietnam. This was led by Joseph Donovan and colleagues. and any folks that have read clinical trials on tuberculous meningitis are probably very familiar with this research group out of Vietnam. and so. I picked this one because I thought it was very forward thinking into precision medicine and how we have to use more information about the immune response to better understand, treatment response, ultimately and, clinical outcomes.
so for TB meningitis, which is the most lethal form of meningitis, there needs to be strides that are made in improving clinical outcomes. And one of those, interventions or strategy has been using corticosteroids in these patients. So suppression of the harmful intercerebral inflammation that you get with TB meningitis, the problem is that there's a lot [00:48:00] of inter-individual heterogeneity in that, inflammatory response, and some of the early work done by this group.
Nav: So in this leukotriene a four hydrolase , so this LTA four H, there's genotypes. So there's a TT genotype, a CC genotype, and a CT genotype. And those sort of vary in their inflammatory response. So the TT genotype is thought of as a hyper inflammatory genotype in terms of the host immune response.
So the question is really for the other genotype, the CT and the CC genotypes, does dexamethasone have an impact? Is there a clinical benefit to it? And then on the flip side, which is why I appreciated Emily's. Talk about the phosphomycin part. Is there even potential for harm, in giving dexamethasone in this subgroup based on genotype?
And so that was what this group of researchers are trying to understand. and they've seen this in some models and they've seen this in [00:49:00] analysis of, previous, studies. they did this clinical trial in patients without HIV and they stratified based on their genotype. And this is sort of the first randomized control trial for TB meningitis stratifying by this genotypic marker for inflammatory response with that, which I thought was really fascinating to see.
their primary objective was sort of twofold. They wanted to know if placebo was non-inferior to dexamethasone in patients without HIV. All of this is in patient with HIV, and they wanted to see is it non-inferior in the CC and the CT genotype. In addition to just, you know, standard anti-tuberculosis treatment.
Nav: essentially if placebo is non-inferior to dexamethasone, then it's showing that it's really little reason to give dexamethasone in this genotypic subpopulation of the CC and the CT group, right? If it's non-inferior on the other end of it, because there might be some evidence of harm with [00:50:00] dexamethasone in this population.
They also wanted to look at superiority of placebo to dexamethasone in clinical outcomes because that would sort of solidify that there is harm from dexamethasone done in this group if placebo is actually superior in terms of clinical outcomes. So it was this kind of two phased. Non-inferiority trial and a superiority trial to look at the difference between dexamethasone and placebo in this genotypic subgroup.
So I thought that in and of itself was pretty innovative and, fascinating to see how they designed this trial with all this, being informed by these previous trials in this genotypic, immune response.
The punchline really for, this, is that there was no evidence of harm for dexamethasone and the placebo is not non-inferior. So that's saying if you put all this together, you should continue to use dexamethasone based on guideline recommendations regardless of this LTA four H genotype.
[00:51:00] there's still, modest benefit, but regardless of the modest benefit, given it being such a high mortality disease, this supports just continued use of dexamethasone, and that's independent of the genotype. And keeping in mind that you don't check or see that genotype in routine clinical practice.
And that's part of the point of this is to see if there is harm, or if there was clear benefit in understanding the use of dexamethasone in these specific genotypes. Should that start to become something that gets looked at in clinical practice to make the stratification clear, but based on these trial results, it doesn't really provide much information to know the genotype.
To actually influence clinical management. and so dexamethasone should be given regardless of what that genotype is, placebo was not non-inferior. And, it didn't also demonstrate superiority, but it took me some time to wrap my head around these sort of dual findings.
when you see that. So, I'll pause there for a second. I'll describe a little bit more of the trial as we go, [00:52:00] but just to get right to that finding, what do you, think when you hear those types of findings?
Emily: so to be not non-inferior, if we break it down, you would need first to set your non-inferiority margin.
and then with your result and its confidence interval, it didn't fall into the non-inferior range. Then look separately and you have to say, okay, for it to be superior, whatever result you have also has a confidence interval around it. And that confidence interval would need to not cross over one.
So you could be both not non-inferior and you could also be not superior and you have a confidence interval and a finding that falls somewhere in the middle.
Nav: That's exactly right. So for folks, I would encourage that are wanting to see this visually. If you look at figure three in the paper, that's exactly what it's showing.
So, just to put some numbers on this qualitative verbal, result that I was presenting, the hazard ratio was 0.99. The [00:53:00] noninferiority margin was, 0.75 of the confidence interval for the hazard ratio and the lower bound just. cross that, and if you look at the figure, it's like basically on, that because the confidence rule spans from 0.75 to 1.31.
And so it's like right on there for this specific CC and CT genotype. And then, it's basically smack dab on one and then the confidence interval spans and just crosses that non-inferiority margin. So exactly what, you described, Emily. So it was deemed both non, non-inferior and superiority was not demonstrated in this, specific, randomized group of the CC and ct.
So that's why they're saying that, dexamethasone has a modest benefit and should be used. Regardless of this genotype. And one other thing I'll say is they had this pre-planned, this a priority planned [00:54:00] individual participant meta-analysis where they took this previous trial that they did in 2004 as published in New England Journal of Medicine that showed dexamethasone had a mortality benefit.
And then they took this trial because this was all done at the same hospital. There was a lot of similarities in the design. And so they did this individual level meta-analysis it had, over 1100 adult patients. and this also just reinforced that, finding that, dexamethasone has a modest benefit in this population of those without HIV, and should continue to be used.
But, I thought it was very interesting 'cause this is the first genotype stratified, randomized control trial. Very much infor being informed. The whole design of it is being informed based on what they're learning about the inflammatory response and how there's genotypes that are associated with that.
The thing is, this was all done in Vietnam. That genotype, you can see more of a spread in that genotype versus what they point out in the study [00:55:00] that this hyper inflammatory genotype, the TT genotype, represents less than 5% of some African populations. And so there's a lot more work to be done to understand the genotype distribution in different populations where TB meningitis is a high risk or, better yet, is in a high TB endemic zone.
So, this was also if, folks were at ESCMID Global Last year. the senior author presented, some of this unpublished data. And so it's just very cool to see it in its final publication. So I thought it was an interesting trial, even though it's not sort of practice changing in the way that there was some major finding that showed this genotype is worthwhile of, finding in a clinical practice and then therefore changing how you use Dexamethasone or not.
But I thought it was very innovative in terms of the design and even just what the setup was. So I'm not sure if you guys have any other thoughts or comments on that.
Yeah, just one reflection from me is that we talk about precision medicine sometimes, and that's been reserved for [00:56:00] cancer treatments for a long time.
Thomas: And I think we're seeing more and more of that in infectious diseases. And I think, like you said, it's just trying that out. See if the genotype of the patients. Should be used in clinical practice to stratify treatment. I think that's a step forward and I think we can probably do much more of that, especially for the more severe or long-term infections, obviously, to try to be more individualized in our treatments.
Emily: Yeah, this study really got me thinking. We're seeing, more use of steroids and pneumonia, for example, and you wonder why some studies are positive and some studies are negative and, you know, there's been lots of discussion around why that's the case. But, you know, one thing to think about is maybe the genotypes of the patients involved in the study differ.
am running a trial of how long to give steroids for, patients with pneumocystis pneumonia. So I think that that's interesting. to think about the genotype. It's tricky of course, because you can't necessarily access that very quickly in clinical practice, but, I think [00:57:00] studying it is, really interesting.
And then, you know, once I suppose if you prove that there is a benefit, then you work on getting access to it in clinical practice.
Nav: And I thought it was interesting too, this trial versus the 2004 trial. there, there are just some differences in. The amount of patients had a definitive diagnosis of TB meningitis, their severity grade.
So there are changes that just happen over the years in terms of diagnostic evaluation, in terms of supportive care, in terms of the severity earlier diagnoses and catching these patients earlier on in therapy and, that also influencing, having a better outcome because the mortality overall was, lower, relative to the, previous trials.
And I think that speaks to other, things that might not be fully quantified in terms of just, the earlier diagnosis and then supportive care that they see. clearly there are some differences in that previous trial, but, I think this is the way to go even though.
This didn't show, don't give dexamethasone as a specific genotype from this [00:58:00] trial. I think this is the right direction to move, as we learn more and are better informed by immune responses in these patients to understand how that'll impact their clinical outcomes ultimately.
Emily: Okay. I think we'll wrap it up there. thank you so much to my two co-host Nav and Thomas for joining me for this inaugural quarterly Catchup episode. We hope you liked it. I'll be interested to see with all of this great discussion, if we'll be able to cut the episode down to 45 minutes. That was our aim, so we'll see how we do there. these will be coming out every three to four months to keep you on top of some of the latest infectious diseases and microbiology science. Thank you for listening to Communicable the CMI COMMS podcast. this episode was edited by Katie Hostettler and peer reviewed by Connor Prosty of McGill University Montreal, Canada.
Theme music was composed and conducted by Joseph McDade. The executive producer of Communicable is Angela Huttner. This episode will be citable with a written summary referenced by A DOI in the next eight weeks, and any published literature that we [00:59:00] discussed today can be found in the show notes. You can subscribe to Communicable wherever you get your podcasts, or you can find it on estimates website for the CMI COMMS Journal.
Thanks for listening and helping CMI, comms and Esid move the conversation in ID and clinical microbiology further along.