Slide Lecture with Audio
This activity, accredited for nurse practitioners and physician assistants, is based on a 40-minute slide lecture presented by Dr. James
S. Lewis II, PharmD. This program is available as read only and with accompanying audio.
*Scroll to read lecture and/or follow along with audio.
Welcome to the teleconference, "Challenges in the Management of Serious Fungal Infections: Counteracting the Impact of Candidemia in the Acute-Care Setting." This program is sponsored by the University of Tennessee College of Pharmacy and Princeton Media Associates, and is supported by an educational grant from Pfizer Incorporated.
I am Dr. James Lewis, Infectious Disease Pharmacy Programs Manager at the University Health System, San Antonio, Texas, and the University of Texas Health Science Center at San Antonio. I will be your presenter today.
My goal in speaking to you today is to, in detail, review the drugs currently available for Candida, talk about a lot of the current clinical trials that have been recently presented in the arena of candidemia, and then also, talk about future directions in this important disease state. Please remember that to receive continuing education credit following the conclusion of this program, each participant must complete the post-test and evaluation form located in the back of the program booklet. These forms can easily be completed online at www.princetoncme.com.
Moving on to Slide 2, the learning objectives for today are, first and foremost, to list the factors that contribute to the prevalence of fungal infection; second, to summarize the burden of candidemia in the acute care setting; and thirdly, to assess the efficacy, adverse event profiles, and tolerability of available agents for the treatment of candidemia. Also, when you come away from this program today, we hope that you'll be able to develop a treatment algorithm that would allow you to figure out when certain drugs should be plugged into certain patient subsets.
So, what we're going to cover going forward is we're going to start with epidemiology, talking about what's going on in the world of Candida infections in your hospital. From there, we're going to move to susceptibility issues, really talking about what the problems are with susceptibility in Candida species, and where those become clinically relevant for you and the patients that you're helping to take care of. Then, we'll move into where we kind of stand in 2006 with antifungal classes. We're going to talk about echinocandins, specifically, because that's really been where a lot of the recent data have emerged.
We're also going to talk about the old reliable drugs, namely the azoles. Fluconazole is still a wonderful agent. We'll talk about where that belongs in 2006. We'll also touch briefly on the newer azoles, voriconazole and posaconazole. And, finally, we'll wrap up talking about the polyenes, good old amphotericin B and its lipid cousins, and where those really should be niched in managing Candida infections in 2006. Finally, what we'll do is wrap up looking at specific species of Candida and talking about what the treatment-related issues regarding those species may be.
Now, when you look at the impact of candidemia in the United States, it really is pretty impressive, and most people don't have a good appreciation for just how important this disease state is. It's the fourth most common bloodstream isolate in US hospitals, and if you look at it, it's clearly the leading fungal pathogen in US hospitals. It's also important to note that some series have moved it as high as number three, as far as bloodstream isolates.
The attributable mortality, according to Zaoutis and colleagues, published in Clinical Infectious Diseases late last year, is about 14.5%. Though, if you really dig through the literature, what you'll find is that range can vary all over the place, and that's probably a low estimate. What becomes important to your pharmacy budget, to your hospital administrators, and to your patients, is that when a patient develops candidemia, you see an increase of about 10.1 days in length of stay, and that's not good for your DRGs, that's not good for your payors, and it's certainly not good for your patients, who are going to see about a $39,000 to $40,000 increase in charges on their hospital bill. And we all know that the longer sick patients hang around the hospital, the more likely it is that they're going to have a problem.
Now, moving on to the next slide, you see here that the thing that you can't do anymore is just say that it's Candida albicans, because C albicans is clearly not just the only player anymore. Back before the advent of fluconazole, it was true that C albicans was probably representing a larger percentage of Candida infections, but now, if you direct your attention to the fourth bar over on this graph, what you see is that Candida glabrata is now representing 20% of US bloodstream isolates.
Now, this is a strikingly eerie graph for me personally because, if you overlay our data from University Hospital here in San Antonio, what you see is that our numbers almost look exactly like this data set from Peter Pappas in 2003. If you go to some places that take care of more cancer patients, and take care of more solid organ transplant patients, sometimes you'll see an even more dramatic shift towards C glabrata and, that bar on the far right, Candida krusei, which can really represent a problem for you in some of your cancer institutions, and we'll talk a little bit later about why that's potentially problematic.
The next slide gives you a table that I have mercilessly adapted from the Infectious Disease Society of America Guidelines in 2003, because I really think that it does a nice job for you. And I think that this is one of the data sets that I would really like to see you walk away from this program hopefully having a better understanding of.
C glabrata, the fourth line down, is where I really kind of want to first start off directing your attention. If you look at susceptibilities for C glabrata, what you notice as you move across this table is that a lot of the azoles appear to have problems. Fluconazole is susceptible dose-dependent to resistant, itraconazole is the same, voriconazole is susceptible to intermediate. Honestly, I would expect posaconazole to look very similar to voriconazole with it being susceptible to intermediate. The problem that we have run into, at least here in San Antonio, and this has been mimicked a little bit in the literature out there, is that when you have fluconazole-resistant C glabrata, oftentimes it is pan-azole resistant, and so I don't know that the newer azoles, even the voriconazole and posaconazole, are gong to buy you a whole lot for fluconazole-resistant C glabrata. Even if you move all the way to the right on the C glabrata bar, you'll see that amphotericin B is into the intermediate range. Tom Walsh and others have been suggesting for a couple of years now that C krusei and C glabrata may have some MIC creep going on with amphotericin B and are recommending that we potentially push the dose up a little bit when managing those 2 species of Candida with amphotericin B.
What then I want to really make you aware of here is looking under the candins column, and when I say "candins," I'm talking about the echinocandins — the three that we now have FDA-approved being caspofungin, micafungin, and anidulafungin — and you can see that all three of those drugs, you've got a susceptible for almost all strains of C glabrata that you'll run into. And so I think you're starting to see here that there are clear differences with regards to the various species of Candida with regards to what drugs they're susceptible to.
So, what I want you to walk away from this with is on the next slide — this quote again stolen from the IDSA guidelines — is that you need to know the species of Candida you're dealing with. It's no longer acceptable for your microbiology lab to tell you things like, "Candida not albicans." They need to give you a precise idea of what it is that you're dealing with, because the susceptibilities are so important for you. If you back up to that last slide, and you start to look at Candida parapsilosis, for example, you see that the echinocandins potentially have a problem there. You look at Candida lusitaniae, the bottom line. You can see that amphotericin B has a problem there. So, again, it's critical that you as practitioners, who are helping physicians make anti-fungal decisions and are helping to take care of patients, you need to know what you're treating. There is no substitute for good microbiology, and it's critical that your microbiology lab be telling you what species of Candida you're dealing with.
The other problem that we really are currently facing with Candida infections is that we are somewhat limited because of the diagnostics that we have available to us. Historically, the gold standard here has really been a positive culture from a sterile site, with evidence of histopathologic involvement, and the problem with that is that blood cultures, by default, have really become kind of the gold standard there. But regardless of what series that you look at for how good blood cultures are for diagnosing Candida, what you quickly discover is that these numbers aren't particularly impressive, with even some of the newer blood culture systems only being about 70% sensitive. So, I would caution you against completely waving Candida off as a diagnosis because of negative blood cultures in patients who are colonized at multiple sites, who have evidence of problems, continued fevers, continued white counts on broad-spectrum antibiotics. I really think that we need to be careful that, just because a blood culture is negative, we're not necessarily saying, "Oh, no. Candida is not even potentially an issue here."
There are some exciting, new diagnostic tests on the horizon that may offer us an improved ability to pick these patients up early, and I will draw your attention to a couple of very nice papers in Journal of Clinical Microbiology in the past several months — the first having to do with the beta-D-glucan assay, which in this paper published by Pickering and colleagues was extremely sensitive, but there were some issues with how specific it was. And so, I think we have a little bit of concern there, especially with the fact that there appeared to be some interference in patients who had gram-positive organisms in their blood at the time that these assays were performed. So, I think we're getting there, but I still don't know that the beta-D-glucan assay is totally ready for prime time yet.
The real time PCR stuff that was presented by Tom Walsh's lab, or Tom Walsh was the co-author with in this paper from Journal of Clinical of Microbiology, again looks promising, but again these are a lot of home brews with really no commercial systems available that you or your clinical lab can walk out and readily buy. So, again, I think we're kind of stuck for the time being, but we're seeing, I think, that the diagnostics are slowly starting to catch-up.
And I really think that with this being such a prevalent problem, being the third or fourth most common bloodstream isolate, that the diagnostic manufacturers will hopefully wake up to the fact that this is an important market financially for them in the near future, which hopefully will drive a lot more development and give us improved diagnostics. Because one of the limitations that we're really facing with these limited diagnostics is that we're probably over-treating people we really don't need to be, and exposing them to some unnecessary risks with anti-fungal therapy.
The next hot button issue that I want to talk about, because I know that this is big for some pharmacists out there, is prophylaxing patients in the intensive care unit, and it's really kind of comical. If you go and look at this on MedLine what you find very quickly is that, I think, there are every bit as many, if not more, meta-analyses on this topic than there are actually good papers. And what I've tried to do on this slide is to really direct you to some of what, I think, are the more important recent papers on this topic, and so that's why there are so many references sitting for you at the bottom of this slide.
So, the question we really face is: does prophylaxis decrease infections? And I think that we can wholeheartedly say from the data that's available, yes, it does decrease infections, especially with C albicans in patients who are treated with fluconazole.
The million dollar question is, "Does it decrease mortality?" And that has really been a much bigger bugaboo. We know that it decreases colonization, it probably decreases infection, but mortality has proven exceedingly difficult to really prove that there's a difference with regards to patients who get prophylaxis for Candida infections.
The other thing that is, I think, a very legitimate concern is that if you start sprinkling anti-fungal all over colonized patients, or even high-risk patients in your ICUs, do you start to select out less susceptible Candida species as a result? I really wonder whether or not places that use low-dose fluconazole are kind of doing this with some of the C glabrata that they see show up. If you put low-dose antifungal out there, an organism that is less susceptible to that antifungal will show up. There are also the tolerability issues. Though fluconazole is an extremely well tolerated drug, it does still have drug interaction potential, it does at times still have LFT issues associated with it. And though fluconazole has gone generic, and is now much less expensive to the tune of maybe $20 a day for 400 mg IV, if you move into other agents, such as echinocandins, you can become very cost prohibitive very quickly. And Golan and colleagues in Annals of Internal Medicine late last year, I think, did a very nice job of looking at this. And though it's a fairly complex paper, I would encourage a lot of you to really go and take a look at this paper if prophylaxis in the ICU is a particular interest of yours.
All right, so that's a little bit of the background, and so hopefully I've given you a sense of what's important with Candida, that microbiology is key. You need to know what you're treating, kind of where we stand with regards to susceptibilities for the different Candida species, where we're going with diagnostics, and what the limitations are for some of the prophylaxis studies at this point.
So now, since we're all pharmacists, let's move into our favorite part here and talk about the drugs specifically.
Let's start with the echinocandins on Slide 11 and really the benefits of these drugs. These are nice drugs. I mean, there's just no way to really debate this. They're rapidly fungicidal against most Candida species. You put them into a test tube with Candida, and you pretty much hear the Candida scream and die, and we ID people get really excited about that.
So, the next thing you've got, though, is a very well-tolerated drug, so you've got a cidal drug like amphotericin B, that unlike amphotericin B is extremely well-tolerated. It's pretty darn hard to hurt somebody with an echinocandin unless you decide to try and give it IV push, and you get a nice histamine reaction out of your patient. The kinetics allow for once-daily dosing, because most of these drugs have a half-life on the range of 9 to 15 hours. They're really not significantly metabolized by the cytochrome P450 system. There is very minimal involvement of P450 at all, especially with anidulafungin. And so the drug-drug interaction compared to what we're used to having to deal with with the azoles are really insignificant, and I think that's really one of the things that has made these drugs so appealing to clinicians.
And then last, but certainly not least, is the fact that you have really no cross-resistance with the azole antifungals. We see so much fluconazole go out the door, and there's been so much buzz about fluconazole resistance with Candida, that it's really nice to know that if you have somebody with Candida in the blood, who is crumping, who has been on fluconazole, and you don't know the susceptibility, that really you don't need to sweat echinocandins being cross-resistance with azole resistant-organisms. So, I think that's really a strength of these compounds, especially in critically ill patients who have seen prior azole therapy.
Now, when you look at the echinocandins, there are some limitations there. I don't want you to think that these are the end all, be all drugs out there, and first and foremost, the limitation is the spectrum. When you look at them, they've got great Candida and Aspergillus activity, but that's really about it. There may be some endemic mycoses activity, but I think, at this point in time, that's very preliminary data, way too early to be encouraging use for those indications, and so I would really stay away from them for things like histomycosis and coccidioidomycosis, and there is no activity whatsoever for Cryptococcus neoformans or zygomycetes.
And where this really becomes important is that when your microbiology lab calls you and says, "We have yeast in the blood," and if you knee-jerk to an echinocandin immediately without knowing whether or not that's Cryptococcus, you could completely miss, and so you have to be very careful and realize that you need to know what you're treating when you're using an echinocandin. Again, another place where there is absolutely, positively no substitute for good answers from your microbiology lab.
The other down side with this class of drugs is that there's no oral formulation available. These drugs are massive molecules, and it's been tried for years to get them into an oral formulation, and nobody has really been able to get the bioavailability up above 1% to 3%. So unless some wizard somewhere cooks up something really cool, you're probably never going to see an oral formulation for these drugs.
The other thing that has been a real limitation for us, especially from a pharmacy budget standpoint, is cost. There is extreme wallet toxicity with caspofungin sitting at about $300 a day for the last 5 years. We've started to see that come down a little bit as micafungin has entered the market. Micafungin is sitting at about $180, although with the newer contract it may be down into the $160 range for some of you. And so, I think that that has really put some pressure on the manufacturer of caspofungin to start looking at coming down on that price, and you've seen it start to come down some. Now that anidulafungin is here with 100 mg of anidulafungin going for about $180 a day, I really expect that you're going to have to see the manufacturer of caspofungin move to get into this range, otherwise I think they're just going to be completely out of the ballgame.
So, we talked a little bit about the strengths and the weaknesses of the echinocandin class. Let's move into the clinical data, because that's really where the rubber meets the road, and there's been a lot of exciting developments in this arena in the last 6 months.
Let's start with kind of the classic paper for this group of drugs, and that's the Mora-Duarte paper from New England Journal of Medicine in 2002 that was the caspofungin candidemia data. You can see from this slide on the table that caspofungin went head-to-head with amphotericin B, and at end of IV therapy in the Modified Intention To Treat group, which is what MITT stands for there, you can see that, you're looking at the slide going, "I'm looking at this, and it looks to me like caspofungin did better than amphotericin B." However, it's critically important that you realize with this paper that toxicity equals failure, and do you think one of those drugs was slightly more toxic than the other? When I think about conventional amphotericin B, I consider it much more toxic than caspofungin, and that's exactly what you saw with 3 patients getting toxicity in the caspofungin arm and 19 in the amphotericin B arm. Remember, this was conventional amphotericin B, not one of the lipid formulations. So, you had a significant difference in toxicity that really accounted for all of the difference that you saw here.
So, the take-home on the study is that both drugs work for Candida, however, conventional amphotericin B is much more toxic, and hopefully that's not too much of a surprise to any of you listening today. I think we all know that conventional amphotericin B can be pretty rough.
So, let's talk about some of the new data now, all right, and this is not even out in peer-reviewed form yet. This was presented at the ICAAC meeting in December of last year, and it's the micafungin versus liposomal amphotericin B for candidemia study that a lot of us had kind of been sitting around waiting on. And this was a double-blind, randomized, non-inferiority study — pretty much everything you want to see. Micafungin 100 mg/day, head-to-head with liposomal amphotericin B at 3 mg/kg/day. The reason for the 3 mg/kg dose is that this was a German study, and that is the European indicated dose.
The neutropenic patients were allowed into the study; I think this is a strength. And again, 50% ICU patients in both arms. So, you had sick patients, potentially neutropenic patients. I think that this makes for a very robust data set.
When you move on to the next slide, you see that treatment success at the end of therapy in the intent to treat arm is about what you would expect, roughly 70%. This fits nicely with what we know for most other antifungal agents, and the take-home here is that micafungin works as well as liposomal amphotericin B, and that's micafungin at 100 mg/day.
Now, you go to the adverse events slide and you look at this, and again you go, "I don't really see anything that's too surprising here." There were more creatinine increases in the liposomal amphotericin B arm. Not too surprising, given that it's an amphotericin B product. Also, this back pain thing, and we see what we like to call the heart attack side effect with liposomal amphotericin B, which is where patients get this crushing chest pain, or they'll get back pain. And so, I think that's kind of what you're seeing here. This is almost a hallmark of liposomal amphotericin B, and it's somewhat unique to that compound for an unknown reason.
But, okay, so I look at the side effects rates here and really there's not too much surprising there, except for maybe that the LFTs look like maybe there was a little bit more of a blip in the micafungin arm there, but it didn't reach statistical significance.
From there, let's talk about the other big study that was presented at ICAAC last year, and that was the anidulafungin candidemia study where anidulafungin at 100 mg/day went head-to-head with 400 mg of fluconazole. This was the basis for the package insert data that you see in the anidulafungin package insert right now, and hopefully, this study will be published in peer-reviewed form here very shortly, but it was a randomized, double-blind, again, non-inferiority study. Eighty-nine percent of patients had candidemia only, meaning they didn't have evidence of involvement of other organs.
The patients were stratified for APACHE II score, so they looked at the really sick ones versus the not-so-sick ones, and again neutropenic patients were allowed. There were 10 days of IV therapy in either arm. So, you got a minimum of 10 days of anidulafungin, and then you could be switched potentially to oral fluconazole in either arm. So, one of the nice little things that this shows us is that you can use 10 days of echinocandin and then switch patients to oral fluconazole if the susceptibility of the organism and the stability of the patient permit that.
You look at the success rates on the next slide, and I'll tell you what. This got my attention in a big hurry. I'm a big fluconazole fan, because it's generic, because I've seen the drug work, because I know how well-tolerated it is. I like fluconazole, I like it a lot, and so it got my attention when something appeared to have beaten fluconazole in candidemia. You look at these confidence intervals on the far-right side of this slide, and you can see that they look like they reached statistical significance. You can see that the delta, or the difference at the end of IV therapy, was 15% in favor of anidulafungin. At 2 week follow-up, 15% in favor of anidulafungin. Even out to 6 weeks- you almost maintain statistical significance even out to 6 weeks. So, though we have not seen this data in full peer-reviewed form yet, I think that this data strongly hints that maybe anidulafungin has beaten fluconazole, and I do believe that the FDA even allowed some superiority verbiage into the package insert. And so that would really be the first time that we've seen this.
So, I think this is an extremely exciting data set, which really potentially changes the way that we have been approaching candidemia, and I would argue especially in your sick patients, I think that now with this data here you're probably obliged to start with an echinocandin up front, especially in somebody who is critically ill and hemodynamically unstable.
So, we've looked at this clinical data here. It looks pretty darn impressive. I think that we know that echinocandins are very good at killing Candida. Now, there are some additional considerations here, before you think that this is just turned into a giant echinocandin party. And the first thing is that, as with any other antifungal, when you put it out there, you're going to start to figure out the bugs that are less susceptible to the drug. And for echinocandins, and that includes caspofungin, anidulafungin, and micafungin, it appears that C parapsilosis may be a bit of an Achilles' heel.
This was borne out by a paper in Antimicrobial Agents in Chemotherapy from Moudgal and colleagues, and also a presentation that Mahmoud Ghannoun made at ICAAC in December. The multi-echinocandin resistant parapsilosis in Antimicrobial Agents in Chemotherapy in 2005 was actually an endocarditis in a patient who developed echinocandin resistance on therapy, and I think that that's very important to note, because parapsilosis runs high MICs at baseline to a lot of our echinocandins.
Mahmoud Ghannoun presented some data from a burn unit in Detroit, where they saw 22 burn patients from June through November of 2004. Nine of those 22 were colonized with C parapsilosis, and these MICs down at the bottom of this slide are high. Normally, what you would see is a 1 or a 2 for C parapsilosis. When you start getting up into 8 and 16, that really is a problem.
One of the things that's interesting here is that in this data set with this group of organisms, anidulafungin appeared to have held its activity at an MIC of about 2, and that may hint that there are some differences within the group, and that maybe not all of these echinocandins are the same.
The other thing, before I go any further, is also I want to point out, going back to the Mora-Duarte study, the caspofungin candidemia study, if you look at the persistent candidemias in that study, especially in the caspofungin arm, even though there are only a few, it will appear that there might have been a problem with C parapsilosis.
The fluconazole versus anidulafungin study, again small numbers, but it appeared that fluconazole might have beaten anidulafungin for C parapsilosis. So I think, again, it's going to be increasingly important for you to know what species of Candida you're treating and to realize that maybe certain bugs are better treated by certain drugs.
So, let's move on to the old familiar compounds, namely the azoles, and hopefully, you guys have seen enough fluconazole, itraconazole, and voriconazole that you probably have a pretty darn good sense for what the mechanism of action is, and that they all pretty much work the same. They inhibit a cytochrome P450 lanosterol 14-alpha demethylase, basically shutting down ergosterol synthesis.
Now, the problem is that you all have cytochrome P450 systems, I have a cytochrome P450 system, and the fungi have a cytochrome P450 system. So, while the azoles are off shutting it down in fungi, they're also inhibiting it in your liver and in your gut, and that's really where all the drug interaction problems come from.
The other thing that's interesting about these compounds is they get kind of, I think, an unfairly bad rap for Candida, because they're considered static. I mean, it is very, very hard to prove what static means clinically in the infectious disease literature. Maybe outside of endocarditis and meningitis, and perhaps osteo, although we can debate that. Static drugs, in a lot of series, appear to do just as well as cidal drugs, and so I have a hard time getting fired up about the static versus cidal, but some people do.
The P450 drug interactions — look at it this way; it's great job security for us as pharmacists. I spend a lot of my time trying to keep my providers out of trouble with voriconazole drug interactions. So, I look at this as kind of a blessing and a curse. The oral formulation, though, is really the strength of this group of drugs, especially for fluconazole and voriconazole, where you have great oral bioavailability on these compounds. It's very nice to get people out the door on these oral formulations.
The other thing is — especially with fluconazole — it's extremely well-tolerated. I think as you get into voriconazole, it gets a little trickier. It's a tougher drug to manage, and itraconazole is probably somewhere in the middle but more towards voriconazole. I think it remains to be seen how well-tolerated posaconazole will be, because we've only had the oral formulation available to us thus far, and haven't gotten high levels into the blood yet with an IV form, although that will hopefully be forthcoming.
Now, when you start to think about azoles and drug interactions, what I've done on this slide is really kind of list some of the biggies. For your solid organ transplant patients, you really have to remember they're immunosuppressive. When you start an azole you usually have to cut back, but when you stop the azole, you have to remember to bring their immunosuppression back to baseline, otherwise you risk rejection of the graft.
Rifampin and rifabutin are just a royal pain in the backside, and are really difficult to deal with with regards to your azole antifungals, voriconazole potentially even being contraindicated with regards to being used in conjunction with rifampin and rifabutin. Your HIV medications are very problematic, and unfortunately, this is a population that tends to require a lot of antifungal. So, when you see HIV patients getting started on an azole, you need to be very cognizant of what is going on with their HIV therapy.
Anti-seizure medications, again, you can see various things happen because of the complex interactions, especially between phenytoin and the P450 system. And then you throw some inhibition in there with an azole, and things start to get really tricky.
The other group of drugs that honestly scare me are a lot of these chemotherapy drugs, where you start to combine them with azoles, and we just don't have a real good sense for exactly what's going to happen, especially the vinca alkaloids being contraindicated with certain drugs. I think we have to be very, very careful, especially with newer antifungals, to try and make sure we have a good sense of what these oncology drug interactions look like.
And then the ubiquitous HMG CoA inhibitors, you're going to see potential interactions there with pretty much all of them except pravastatin, but with the big move to drugs like cerivastatin and simvastatin and, of course, atorvastatin, which is currently, I believe, the market leader, you've got to really be careful because you can shut down the metabolism of these drugs, and the next thing you know somebody has rhabdo on you.
All right, so let's talk specifically about some of the antifungals here, and I will first talk about fluconazole, which is clearly the most used systemic antifungal currently on the market. The benefits of this drug I think I've touched on a little bit already, but I'll go over them again. First and foremost, I really believe that it's the lack of toxicity that makes fluconazole such a nice drug. I mean, really it's very hard to hurt somebody with fluconazole, even up to into the ranges of 1600 mg/day to 3200 mg/day, not that I'm encouraging those types of doses routinely.
The oral bioavailability of this compound is outstanding. Pam Lipsett at Hopkins has done some great kinetic data. Even in critically ill patients, if you're putting it down tubes they still absorb it, they still get great blood levels, and it's really active against many Candida species, with the exceptions being C glabrata and C krusei. I think that you also have to kind of watch your HIV patients, and any patient who has bathed in prolonged fluconazole prophylaxis or treatment, you have to be a little bit concerned about the emergence of resistance. But for the most part, it's really only C glabrata, and even then, that's maybe only 30% of all your C glabrata isolates that are going to have problems with fluconazole.
The limitations, all right, the biggest problem as I see it with fluconazole is that people still aren't getting the dosing right. We really need to be using a minimum of 400 mg of fluconazole a day to treat Candida infections. If you look at the IDSA guidelines, they do a very nice job of spelling it out that it needs to be 6 mg/kg/day minimum in adults, and that works out in the 70 kg adult to be about 400 mg a day. I think one of the biggest disservices that we have done to fluconazole is underdosing it, potentially allowing or selecting out for some of these C glabratas by just using these low levels of fluconazole. So please, if you're going to use fluconazole in your patients, make sure that you're dosing it appropriately.
From there we move into, again, something that is near and dear to my heart, and this is data that we presented at IDSA in 2004, and we're hoping to present again in an updated form at IDSA in 2006, having to do with fluconazole distifusion testing. And this is a way that you and your lab can really quickly come up with answers for especially C glabrata, because what we were seeing was that there was a lot of concern about C glabrata resistance to fluconazole, and that was driving a lot of echinocandin use, and at the time we were spending about $300/day for caspofungin. And so, we went around and looked for a way that we could potentially come up with fluconazole susceptibility answers quickly. And what we found is that about 70% to 80% of our C glabratas, depending on which 6 months we're looking at, are fluconazole-susceptible. And this has really helped us keep our echinocandin expenditures in line, because we've been able to provide our providers with answers for fluconazole susceptibility very quickly, and there are multiple methods to do this. And I just show you this because it's quick, it's dirty, we do it, it's worked well for us.
There are also other systems available, though, that can give you this answer, and I think this is really important, especially if you're seeing a lot of echinocandin use going out of your pharmacy.
So where does fluconazole belong in 2006? That's a really good question. I think where it belongs is that you really have to watch it for C glabrata, especially in patients who have seen prior fluconazole therapy.
The other question that always comes up is, "Was it as good as amphotericin B?" If you go back and you pull some of the early data looking at fluconazole versus amphotericin B, what you always find is that the amphotericin B looks like it's trending towards being a little bit better, but it never reaches statistical significance, and I think that's where a lot of the bias comes from. The providers still feel that amphotericin B is the gold standard and that fluconazole is, in some way, inferior.
The thing that I do think is extremely interesting is the anidulafungin data. I do think that we may have seen the dethroning of fluconazole as kind of your first-line drug, especially in the ICU. I think we have to be very careful there. The one thing I will say, though, is remember that it does look like fluconazole may have done a little bit better for C parapsilosis, and so the question becomes, "Where do I put fluconazole in 2006?"
I think really where this drug sits right now is probably for the patient who is on your medicine or surgery floor, who is not hemodynamically stable, who looks like they're clinically improving, who is able to take an oral medication, and who has what is probably a line-associated candidemia. Those are the types of individuals where you have time to do the fluconazole susceptibility, you can get them started on fluconazole, they're not on death's door or anything like that.
I think maybe where we're not going to use as much fluconazole anymore is going to be in the ICU. I think in individuals who have Candida in the blood, who are hemodynamically unstable, I think that those individuals now, especially because of the anidulafungin versus fluconazole data, probably warrant an echinocandin up front.
And again, I would remind you of the C parapsilosis data and remind you of the importance of making sure that you know exactly which organism it is you're treating.
With regards to voriconazole for candidemia, you can see this study here that was published nicely in Lancet late last year; in that it looks like voriconazole works for Candida, and I don't think that's much of a surprise to any of us. This study, though, was voriconazole head-to-head with amphotericin B that could then be transitioned to fluconazole. So, in all actuality, it was voriconazole versus a couple of days of amphotericin B, followed by fluconazole. And some people have said maybe this really isn't a good voriconazole versus polyene, and you really can't say that voriconazole is equivalent to a polyene.
I think you can kind of interpret the data either way. I will leave that to your discretion. I think what this data shows is that really voriconazole does work for candidemia, but between fluconazole being generic and now the availability of the echinocandins, I'm not sure how much voriconazole we should really be using for Candida anyway.
Talking briefly about posaconazole, at this point, posaconazole will probably be available some time this summer. The spectrum looks very similar to voriconazole with the exception that it does pick up zygomycetes to a certain extent. The limitation of posaconazole is that it's currently only available in the oral formulation, although that oral formulation does taste great, especially if you put it over ice cream.
The other thing is, it may be slightly cleaner than voriconazole with regards to drug interactions, but I'm still being very cautious about that, at this point, because we have not seen this drug in IV form yet, and that's what I'm still waiting on. The oral formulation has been very well-tolerated in our hands, and we've had some great saves with it, although we have not really used it for Candida. And so, I think maybe this drug comes in as potentially part of a prophylaxis-type structure, because it does have the Candida coverage there, very similar to voriconazole, but I don't know that this is going to be a big player in candidemia, per se.
So, moving on to amphotericin B, where does it stand right now? And I would really, if any of you have not read this excellent paper by Louis Ostrosky in Clinical Infectious Diseases from 2003, I would strongly encourage you to pull it out. I think it's a very good paper, and I think it has a lot of important points for us as pharmacists in it.
Amphotericin B, I think, is still the gold standard comparator, and I think that that is very, very unfortunate. I think that the FDA should really allow us to be using lipid formulations of amphotericin B as comparators for studies, because honestly, I'm not going to allow my mom to be enrolled in a conventional amphotericin B study, or a study where conventional amphotericin B is the comparator.
The toxicity, I think, is abundantly clear. You look at the Mora-Duarte study, you look at various papers that are scattered all throughout the literature that patients get into trouble on conventional amphotericin B.
Now, it may be very nice that we can get it for $10 a vial, but I don't know that we're really doing our patients any favors by doing that, and in all honesty, I'm not sure we're doing that. I'm pretty darn sure, as a matter of fact, we're not doing the healthcare system any good by continuing to use conventional amphotericin B. I think maybe in neonates it's okay. Maybe in an HIV patient with Cryptococcus meningitis, who has never seen amphotericin B before and has no other nephrotoxins on board, maybe it still makes sense, but I don't know that conventional amphotericin B remains a front-line player anymore.
The lipid formulations of the amphotericin B, though, I do think remain interesting because of the decreased nephrotoxicity, but remember at the end of the day, they are still amphotericin B products, and I think that with the echinocandins available, with the newer azoles available, and with the track record that fluconazole specifically has for Candida, I just don't know that lipid amphotericin B, even though it is less toxic in the short run, really remains much of an option for Candida.
Now, to kind of wrap this up, let's start looking at organism-specific problems here, and let's start with C glabrata. There's a very nice paper by Borst and colleagues in Antimicrobial Agents and Chemotherapy last year that showed that you can get rapid and stable fluconazole resistance in C glabrata, and this did not take much time at all, and the problem was that it did not reverse when the drug was removed. And so, I think in patients who are receiving fluconazole for proven C glabrata infections, it is wise to continue to keep a good eye on them. Even if they appear to respond clinically up front, I would probably still watch them for the next few days. I think this is one of the hot debates right now kind of in this arena, is where does fluconazole belong for C glabrata fungemia?
The role of voriconazole for C glabrata, I think, is a very good question. My own personal opinion is that, with fluconazole-resistant glabrata, I have never seen one that's fluconazole-resistant that isn't voriconazole-resistant, and so I think if you have a fluconazole-resistant bug, I don't know that voriconazole really buys you anything.
The pan-azole resistance, I think, is a problem with C glabrata. The efflux pumps in C glabrata have proven that they are exceptionally good at pumping not just fluconazole, itraconazole, voriconazole, but they're already pumping posaconazole, and it isn't even on the market yet.
If you're going to use a polyene in this situation, I would encourage you, again, to be using at least 1 mg/kg/day of conventional amphotericin B, or 5 mg/kg/day of a lipid formulation.
The echinocandins, I think, are really the drugs of choice right now for severe C glabrata infections. I will caution you, though, that you need to keep an eye on this. Somebody who continues to kind of show up with C glabrata while on echinocandin therapy may warrant susceptibility testing or MIC testing for that C glabrata, because we have seen a couple of these become resistant to echinocandins on therapy, including one that was presented a couple of years ago by one of our fellows.
Now, moving on to C krusei, this is really more of a problem in cancer centers, especially cancer centers that use a lot of fluconazole prophylaxis, and the reason that it's probably an issue in those settings is that it's intrinsically fluconazole-resistant. Fortunately, this organism is rare in most institutions that are not big cancer centers, and also, it does appear to have a lower mortality risk associated with it than C albicans.
Treatment options, and I think this is one place where voriconazole becomes a very interesting option, because voriconazole is very active against C krusei, and fluconazole has no activity whatsoever. So, I think as far as oral step-down therapy for somebody who has got C krusei, I think voriconazole is a viable option. I think your echinocandins are a very good option here. And again, if you're going to use an amphotericin B product here, remember that there is MIC creep with C krusei as much as there is with C glabrata, and I would be using conventional amphotericin B at 1 mg/kg/day, or lipid formulations of amphotericin B at 5 mg/kg/day.
C parapsilosis, I think, is going to be a bug that we have to keep an eye on. This may be the Achilles' heel of the echinocandins, although I think it's a little too early to pass judgment on that yet. I think it will be very interesting to watch, especially in units that use a lot of echinocandin, to see whether or not C parapsilosis becomes a big player, much as C glabrata has in institutions that use a lot of especially low-dose fluconazole.
The other thing that's potentially problematic for the echinocandins here with C parapsilosis is that it does appear that it's able to become a class resistance problem, as was nicely demonstrated in the paper in Antimicrobial Agents in Chemotherapy last year. The caspofungin candidemia study is nice, but remember the difference between caspofungin and amphotericin B was really due to toxicity there. And I would also want you to make sure that you keep an eye on those C parapsilosis patients, because remember in the caspofungin candidemia study, it looked like maybe that was where the problem was.
The same thing in the anidulafungin candidemia study — fluconazole actually looked better than anidulafungin for C parapsilosis. So, I think we need to keep a real close eye on patients, especially who are being treated with an echinocandin, who have C parapsilosis as their positive organism.
The other thing is the experience in the burn unit, I think, serves as something of a harbinger that C parapsilosis can move around through your unit and may be an infection control issue that could be driven by echinocandin use.
I also want to wrap up here with a little glimpse at something that I found to be extremely intriguing, and this is a paper that was published in Clinical Infectious Diseases this month. And this is a recombinant monoclonal antibody targeting heat shock protein 90 that was found some years ago to be somewhat of a predictor of good response, especially in animal models with candidemia. And what they did, when they looked at this monoclonal antibody in combination with lipid amphotericin B, mostly ABLC, although there was some liposomal amphotericin B in the study as well. These patients all had invasive candidiasis, the vast majority of it was candidemia, and interestingly enough, it looks like the combination of heat shock protein 90 with amphotericin B was better than amphotericin B alone. And I would again draw your attention to this table at the bottom of the graph looking at the confidence intervals in the P values. You can see a P value of less than .001 for complete response comparing heat shock protein 90 plus ABLC or LAMB to using placebo, or amphotericin B by itself, I should say. And again, you move down here to the attributable mortality, and it looks like there was a mortality benefit by adding this heat shock protein 90 monoclonal antibody onto the lipid formulation of amphotericin B.
So, while this is kind of a preliminary study, I think that this is potentially very exciting, given that we still see 30% failure in some of our candidemia studies depending on what drugs we're looking at and whatnot. So, I think this may be a glimpse at some things that are to come. We still need big studies here, and this product is not commercially available yet.
All right, so in conclusion, Candida in the ICU is still a moving target. In critically ill patients, who are hemodynamically unstable with Candida in the blood, I think that echinocandins are probably now the first-line agents. Given their safety, given their tolerability, given their lack of azole cross-resistance, I really think these are your first-line drugs for critically ill patients in the ICU.
The role of prophylaxis in the ICU, I think, is still hotly debated. You either love it or you hate it. There are more good meta-analyses on this subject than there are good papers, and I'm not really sure what the right answer is. I know that we can prevent colonization, we can probably prevent infection, but we're not really decreasing mortality, which at the end of the day is what you want to do when you go out prophylaxing patients. It may just be, though, that we have not found the right subset to look at to study in this area, and I will tell you that doing those studies in an ICU is very, very hard to do. So, that I think is still something of a moving target.
Which drugs when? I'll tell you right now. I've already tipped my hand on this. I think the echinocandins should be your first line drug for critically ill patients in the ICU who have been on prior azole therapy while you wait for an identification of Candida in the blood. Fluconazole still has a role on your general medicine, general surgery floors, patients who are otherwise stable, no prior azole therapy. I think fluconazole is a perfectly reasonable first-line choice in that setting.
Are we seeing a changing of the guard? Absolutely. I think you've really seen amphotericin B relegated to third-line status because of the safety, tolerability, and cidal activity of the echinocandins, as well as the safety, tolerability and oral formulation of the azoles. I really think you've seen amphotericin B pushed to kind of third-line therapy here, and I just don't see a whole lot of role for amphotericin B in the future of managing Candida infections. Now, maybe if you start combining it with heat shock protein 90, who knows? But I think that's a ways off as well.
So with that, I will thank you very much.
