Facing the Challenges of Invasive Fungal Infections: Clinical Updates & Best Practices

Transcription

1 Facing the Challenges of Invasive Fungal Infections: Clinical Updates & Best Practices Faculty: Douglas Slain, PharmD, Chair Melissa Johnson, PharmD Ryan Shields, PharmD Digital Guidebook Jointly provided by ProCE, Inc. and the Society of Infectious Diseases Pharmacists (SIDP), and supported by an educational grant from Astellas Scientific and Medical Affairs, Inc.

2 Facing the Challenges of Invasive Fungal Infections: Clinical Updates & Best Practices Digital Guidebook Invasive fungal infections (IFIs) are life-threatening conditions that require astute diagnosis and management to mitigate the high associated morbidity and mortality risks. IFIs are also associated with a high economic burden, owing to prolonged hospitalization, the need for intensive supportive care, and the use of costly new antifungal agents. Opportunistic IFIs occur predominantly in immunocompromised patients, and the incidence of these infections is on a steady increase, as more and more patients undergo surgical procedures, transplantation, or cancer treatment and are receiving immunosuppressive therapies. Several classes of antifungal agents can be used to treat these invasive infections, but some fungi have developed resistance and no longer respond to standard antifungals intended to eradicate them. New therapies offer the potential to improve outcomes for patients with IFIs, but early diagnosis remains a critical component in the effective management of these deadly infections. Clinicians should be aggressive in seeking a diagnosis in patients who are suspected of having an IFI, and the use of available diagnostic tools can improve rates of detection. As medication experts, pharmacists in health-system settings need to assess antifungal use as part of their antimicrobial stewardship programs. This enduring webcast presents the proceedings from a live satellite symposium held at the 2016 ASHP Midyear meeting in Las Vegas, NV. Faculty will review emerging data and strategies on the diagnosis and pharmacologic management of IFIs, and will use case examples to demonstrate the clinical application of these principles. Learning Objectives The target audience for this activity includes pharmacists and infectious-disease pharmacists in health-system settings. At the completion of this activity, the participant will be able to: Outline the benefits and limitations of available tools for early diagnosis of invasive fungal infections to select appropriate diagnostic methods. Design guideline-based management strategies for invasive fungal infections, particularly for immunocompromised and immunosuppressed patients, including appropriate prophylactic and empiric treatments. Identify the risk and benefits of new therapies for invasive fungal infections to incorporate them into safe and effective treatment plans. Discuss the role of susceptibility testing and therapeutic drug monitoring in ensuring efficacy and safety of antifungal treatments. Faculty Douglas Slain, PharmD, BCPS, FCCP, FASHP (Chair) Professor Infectious Diseases Clinical Specialist West Virginia University Morgantown, West Virginia Melissa D. Johnson, PharmD, MHS, AAHIVP Associate Professor of Medicine Division of Infectious Diseases & International Health Duke University Medical Center Liaison Clinical Pharmacist Duke Antimicrobial Stewardship Outreach Network Durham, North Carolina Ryan K. Shields, PharmD, MS Assistant Professor of Medicine Division of Infectious Diseases University of Pittsburgh Pittsburgh, Pennsylvania This CE activity is jointly provided by ProCE, Inc. and the Society of Infectious Diseases Pharmacists. ProCE, Inc. 848 W. Bartlett Road Suite 3E Bartlett, IL

3 About the Faculty Douglas Slain, PharmD, BCPS, FCCP, FASHP (Chair) Dr. Douglas Slain is a Professor and Infectious Diseases Clinical Specialist at West Virginia University. He received his B.S. Pharmacy degree and his Pharm.D. degree from Duquesne University in Pittsburgh. He completed a residency and fellowship in Infectious Diseases Pharmacotherapy at the Medical College of Virginia (Virginia Commonwealth University) in Richmond. Dr. Slain has the distinction of being a Board Certified Pharmacotherapy Specialist (BCPS) with added qualifications in Infectious Diseases. Additionally, Dr. Slain was selected as Clinician of the Year by the Society of Infectious Diseases Pharmacists in Dr. Slain s clinical practice sites at WVU include the Infectious Diseases Consult Service and the outpatient Infectious Diseases clinic. Melissa Johnson, PharmD, MHS, AAHIVP Melissa D. Johnson, PharmD, MHS, AAHIVP is a Liaison Clinical Pharmacist for Duke Antimicrobial Stewardship Outreach Network (DASON) and Associate Professor of Medicine in the Division of Infectious Diseases & International Health at Duke University Medical Center in Durham, North Carolina. Dr. Johnson received a B.S. in Biochemistry from University of Georgia, a Doctor of Pharmacy degree from Campbell University, and a Masters in Health Science in the Clinical Research Training Program from Duke University School of Medicine. Following a fellowship in Infectious Diseases Pharmacotherapy at Duke University Medical Center, she joined the faculty in the Division of Infectious Diseases & International Health. Dr. Johnson has served as investigator for numerous clinical trials with antifungal, antiretroviral, and antibacterial agents. Her clinical research interests include invasive fungal infections in immunocompromised hosts with special focus on immunogenetics, pharmacogenetics, and pharmacodynamics. She was the recipient of a 5-year NIH/NIAID Mentored Career Award to pursue patient-oriented research in invasive candidiasis, and was co-investigator on an NIH program grant to investigate microfluidic methods of detection for infectious pathogens including Candida spp. She rounds on the Transplant Infectious Diseases Consult service at Duke University Hospital, and precepts PGY2 residents on this rotation. She has been a member of the Duke Antimicrobial Stewardship Evaluation Team since In 2015, Dr. Johnson joined DASON to expand involvement in stewardship efforts as a Liaison Clinical Pharmacist in a network of community hospitals in the Southeastern United States. Dr. Johnson serves as a scientific reviewer for numerous journals, and is Associate Editor of Infectious Diseases for Frontiers In journals. She has been an invited international and national speaker on topics such as antibiotic resistance, HIV, invasive fungal infections, and management of bacterial infections. She is an active member of the American College of Clinical Pharmacy, American Society of Microbiology, and Society of Infectious Disease Pharmacists. Ryan Shields, PharmD, MS Ryan Shields, PharmD is an Assistant Professor in the Departments of Medicine and Clinical and Translation Research at the University of Pittsburgh. He is also a clinical pharmacist in the Division of Infectious Diseases at the University of Pittsburgh Medical Center, where he provides patient care through the Transplant Infectious Diseases service, and the Antibiotic Management Program. Dr. Shields NIH-funded research focuses on antimicrobial drug resistance, including antifungal drug resistance among Candida species. His work in this field addresses the use of molecular markers of antifungal resistance to improve patient outcomes. Faculty Disclosures It is the policy of ProCE to require the disclosure of the existence of any significant financial interest or any other relationship a faculty member or a sponsor has with the manufacturer of any commercial product(s) discussed in an educational presentation. Dr. Johnson is an Author for UpToDate and a Principal Investigator for Charles River Laboratories. Dr. Shields is a Consultant for Cidara and an Investigator for Merck and Theravance. Dr. Slain has no relevant commercial or financial relationships to disclose. 2

4 Accreditation This CE activity is jointly provided by ProCE, Inc. and the Society of Infectious Diseases Pharmacists (SIDP). ProCE, Inc. is accredited by the Accreditation Council for Pharmacy Education as a provider of continuing pharmacy education. ACPE Universal Activity Number H01-P has been assigned to this knowledge-based home-study CE activity (initial release date ). This activity is approved for 1.5 contact hours (0.15 CEU) in states that recognize ACPE providers. This CE activity is provided at no cost to participants. Statements of completion will be issued online at upon completion of the evaluation and post-test with a score of 70% or higher. Proof of completion will be posted in NABP CPE Monitor profiles. No partial credit will be given. Release Date: Expiration Date: This activity is supported by an educational grant from Astellas Scientific and Medical Affairs, Inc. Disclaimer: The opinions expressed in this activity should not be construed as those of the CE provider or Astellas Scientific and Medical Affairs, Inc. The information and views are those of the faculty through clinical practice and knowledge of the professional literature. Portions of this activity may include the discussion of drugs for unlabeled indications. Use of drugs outside of labeling should be considered experimental, and participants are advised to consult prescribing information and professional literature. 3

5 Faculty Douglas Slain, PharmD, BCPS, FCCP, FASHP (Chair) Professor Infectious Diseases Clinical Specialist West Virginia University Melissa D. Johnson, PharmD, MHS, AAHIVP Associate Professor of Medicine Division of Infectious Diseases & International Health Duke University Medical Center Liaison Clinical Pharmacist Duke Antimicrobial Stewardship Outreach Network Ryan K. Shields, PharmD, MS Assistant Professor of Medicine Division of Infectious Diseases University of Pittsburgh Dr. Johnson is an Author for UpToDate and a Principal Investigator for Charles River Laboratories. Dr. Shields is a Consultant for Cidara and an Investigator for Merck and Theravance. Dr. Slain has no relevant commercial or financial relationships to disclose. 2 4

6 Which Candida species is is responsible for for significant increases in in antifungal resistance in in the the past decade? 4 4 Audience Response We re going to begin with an audience response question. The first question is: Which Candida species is responsible for significant increases in antifungal resistance within this past decade? You can see there are four species listed there. We have responses all over the place here, which is always good for discussion. That means that we have a lot we can add. So, the correct answer is actually Candida glabrata. And what we re going to find out is that it s not just one class of drugs that we re seeing increased resistance in over this past decade. Which antifungal class had had the the greatest increase in in hospital inpatient use use during the the past decade? Audience Response One more question: Which antifungal class has the greatest increase in hospital inpatient use during the past decade? Think about the past ten years. Which of these classes has the greatest increase? The correct answer, most of you got: echinocandins. And I ll show you some data that come from the CDC. 5 5 Most Common Opportunistic Invasive Fungal Infections This is a slide that shows you what the most common invasive fungal infections are that we see, largely in hospitalized patients. It shows you Estimated life-threatening Mortality rates rates in in Candidiasis > 400,000 46% 46% --75% 75% Cryptococcosis > 1,000,000 20% 20% --70% 70% Mucormycosis > 10,000 10,000 30% 30% --90% 90% Infection infections/year infected population Aspergillosis > 200,000 30% 30% --95% 95% some amount of incidence based on these lifethreatening infections per year. You can also see mortality rates in the infected population. You can see that the mortality rates have great variance. Some of them get very high. Pneumocystis (PCP) (PCP) Brown GD, et al. Sci Transl Med. 2012;4(165):165rv13. Brown GD, et al. Sci Transl Med. 2012;4(165):165rv13. > 400,000 20% 20% --80% 80% Remember, though, that these are not attributable mortality, but these are all-cause 6 6 mortality rates that are reported here. So, there is some variance. Obviously, the ones that have the greatest mortality tend to be the molds, like Aspergillus and mucormycosis. But in terms of commonality, Candida is by far the most common, or it s the most common that we see; but in terms of life-threatening infections, Cryptococcus actually has quite a bit. 5

7 Distribution Distribution of of IFIs IFIs (%) (%) Distribution of Invasive Fungal Infections by Now, depending on the types of services that Patient Category (PATH Alliance) you offer at your hospital, the types of fungal infections that you re going to see are going to vary. This is some data collected from the PATH system, the PATH Alliance, which is a network of hospitals. And you can see what the variability is between having invasive candidiasis, invasive aspergillosis, mucormycosis, and Cryptococcus. 0 0 Medicine Medicine Hem-Malig Hem-Malig HSCT HSCT SOT SOT Solid Solid Tumor Tumor HIV HIV Surgery Surgery And it s quite different depending on the types Invasive Invasive Candidiasis Candidiasis Invasive Invasive Aspergillosis Aspergillosis Mucormycosis Mucormycosis Cryptococcosis Cryptococcosis of services that you re seeing. Most hospitals Azie et al. Diag Micro Infect Dis. 2012;73: Azie et al. Diag Micro Infect Dis. 2012;73: obviously have surgery, so you tend to see more Candida than you see of the others. Also, in medicine you see a lot more Candida. It s really when you get into the more immunocompromised hosts where you start to see some of the mold infections. And if you re dealing with a significant population of HIV patients, that s where you re going to see a greater percentage of cryptococcal infections. Candida: Common Pathogen in in a Nationwide Collection You ve probably heard the stat before, but of of Nosocomial Bloodstream Isolates (N=20,978) Candida is often referred to as the fourth most common bloodstream isolate in a hospitalized BSI/10,000 % of of BSI BSI % of of BSI BSI ICU ICU Non-ICU Non-ICU Pathogen Pathogen Admissions in in ICU ICU Non-ICU Non-ICU Mortality* Mortality* patient. These are some data from what was Coag-Neg Coag-Neg Staph Staph S. the SCOPE database, and this one is from their S. aureus aureus Enterococcus spp. spp report in But others tend to think that Candida Candida spp. spp the rate of Candida bloodstream infections is E. E. coli coli still probably about this, where it falls into Klebsiella Klebsiella spp. spp about the fourth most common. You can also ** Crude Crude Mortality Mortality Rates Rates Data from a nationwide, concurrent surveillance study (Surveillance and Control of Pathogens of Epidemiological Importance Data from a nationwide, concurrent surveillance study (Surveillance and Control of Pathogens of Epidemiological Importance [SCOPE]). Clin Infect Dis. 2004;39: see they re reporting the mortality rates, and [SCOPE]). Clin Infect Dis. 2004;39: compared to some of these other organisms that we think of being very virulent, Candida can have a rather high mortality rate. But again, I will tell you that this is not necessarily attributable rates of mortality, but people can have mortality based on their underlying disease states. 6

8 Candida Species Causing Candidemia in in North America ( ) C. C. albicans C. C. tropicalis C. C. parapsilosis C. C. glabrata C. C. guilliermondii C. C. krusei krusei Other Other Spp. Spp. In terms of what species we see, this is the breakdown from 2004 to 2008, from a series of hospitals throughout North America. This has not changed that much overall, when you take all comers. Candida albicans is typically about 50% of the pie. It might flux a little bit. The next most common isolate, of course, is typically Candida glabrata. And we can often see things like parapsilosis and tropicalis. Nishikaku AS et al. Curr Fungal Infect Rep. 2010; Nishikaku AS et al. Curr Fungal Infect Rep. 2010; Association Between Hospital Mortality and Initiation of Antifungal Therapy in in Candidemia P=0.169 Mortality (%) 20 (%) 20 N= N= < to to to to > Time Time from from Draw Draw of of First First Positive Positive Culture Culture to to Start Start of of Antifungal Antifungal Therapy Therapy (hours) (hours) Morrell et al. Antimicrob Agents Chemother. 2005;49: Morrell et al. Antimicrob Agents Chemother. 2005;49: The next two slides really emphasize the point about the importance of having appropriate antifungal therapy administered as early as possible in patients that have invasive fungal infections. This is a study that was done that looks at association between hospital mortality in patients that ultimately are diagnosed with candidemia, showing when you diagnose it and start therapy. You can see that the quicker you start, you can actually have lower mortality. There have been several studies that have shown this. Association Between Hospital Mortality and Initiation of Fluconazole Therapy in in Candidemia Mortality (%) (%) N= N= P= Culture Day Day Day Day 1 Day Day 2 Day Day > 3 Days Days to to Start Start of of Fluconazole Garey et al. Clin Infect Dis. 2006;43: Garey et al. Clin Infect Dis. 2006;43: This slide is by one of our colleagues, Kevin Garey, and his colleagues. You can see in this study, specifically with starting fluconazole in patients that had candidemia, when it was initiated on culture day versus day one, day two, day three, how the mortality difference is significant, based on when you started. Ryan will talk to us a little bit about some of the diagnostic tests that we have. And we really struggle now, because a lot of times it s hard to find these isolates to get the drug therapy in there as quickly as possible. 7

9 Percent Percent of of resistant resistant isolates isolates Candida Bloodstream Isolates Showing Fluconazole, Echinocandin, or or Multidrug Resistance This is a slide about some resistance issues that we ve seen in recent years. The blue bar 10% represents fluconazole resistance, the pink or red 10% 9% 9% 8% 8% 7% 7% 6% 6% 5% 5% 4% 4% 3% 3% 2% 2% 1% 1% 0% 0% bar is echinocandins and the green bar is multidrug resistance. That mainly is talking about resistance to both these classes. For years, we always had some rate of fluconazole resistance in particular, and you can see that it s remained Year Year of of Surveillance pretty steady. You can also see that over recent Fluconazole Fluconazole Echinocandins Echinocandins Multi-drug Multi-drug years we re seeing a little bit of an increase in echinocandins. Obviously, that s going to increase our multi-drug resistance. There can be some differences within the azole class, and we ll have some discussion about that. Proportion of of Candida glabrata Cases with This is a slide that comes from another network, Echinocandin-Nonsusceptible Isolates by by Hospital Site* and this is actually put together by the CDC. Now, I want to show that the X axis here is not time. It looks almost like it s over time and something is increasing. This is from a network of hospitals that generated a certain number of positive bloodstream isolates for Candida, and it s just showing you the rate of Candida glabrata * * Among Among hospitals hospitals with with isolates isolates submitted submitted during during the the study study period, period, , , sorted sorted by by that they see, and the rate of drug resistance or proportion proportion of of nonsusceptible nonsusceptible isolates. isolates. This work is written by (a) US Government employee(s) and is in the public domain in the US. nonsusceptibility, specifically to echinocandins. This work is written by (a) US Government employee(s) and is in the public domain in the US. Available at: Vallabhaneni et al. Open Forum Infect Dis. 2015;2:ofv163. Available at: Vallabhaneni et al. Open Forum Infect Dis. 2015;2:ofv You can see, in a network of hospitals that submitted their data and had to have 20 good isolates of Candida glabrate, the rates are a little bit all over the board. That reflects the mix of the hospital. But you can see where it has gotten even as high as 25% and 28%. 8

10 Distribution of Candida Species in in Bloodstream You re probably familiar with some of the data Based on Duration of Antifungal Prophylaxis here. This is just a nice pie chart published in an article in the New England Journal of Medicine. C. You can see that the impact of antifungal C. albicans albicans No No Antifungal Antifungal Prophylaxis Prophylaxis C. C. dubliniensis dubliniensis prophylaxis can actually explain why you C. C. tropicalis tropicalis C. C. glabrata glabrata ultimately might see different species of C. C. krusei krusei C. C. cerevisiae cerevisiae Candida. Because we know that the use of C. C. parapsilosis parapsilosis antifungals, based on duration and also sometimes on dose, could even select certain Kullberg BJ, Arendrup MC. N Engl J Med. 2015;373: Candida species that we might see. And here Kullberg BJ, Arendrup MC. N Engl J Med. 2015;373: you can see that, if there s no antifungal prophylaxis, you can use that to compare if prophylaxis was short-term, less than seven days or greater than seven days, and it can cause differences. Distribution of Candida Species in in Bloodstream Specifically, some people will use fluconazole Based on Antifungal Prophylaxis Agent Used and maybe some other azoles, for antifungal prophylaxis or preemptive therapy; or they can No even use echinocandins. And this is just a No Antifungal Antifungal Prophylaxis Prophylaxis C. C. albicans albicans display to show you how the impact of the C. C. troplicalis troplicalis different preemptive agent might change the C. C. glabrata glabrata C. species that you might ultimately see. We C. krusei krusei C. C. parapsilosis parapsilosis know that a lot of use of fluconazole, for example, will increase things like your percentage of Candida krusei, glabrata, and Kullberg BJ, Arendrup MC. N Engl J Med. 2015;373: Kullberg BJ, Arendrup MC. N Engl J Med. 2015;373: even tropicalis in some reports. Now, caspofungin, which typically you ll see higher MICs to parapsilosis you can see where the parapsilosis distribution is a little higher in the patients that would have received caspofungin. Very briefly, these are the species of Aspergillus that we see in our hospitals. Obviously, A. fumigatus is the most common. 9

11 Antifungal days of therapy/ 1000 patient days Antifungal days of therapy/ 1000 patient days Azole Resistance Frequency in in Aspergillus fumigatus Among And there hasn t been a lot on drug resistance Isolate Submitted for for Susceptibility at at The Mycology Reference Centre Manchester here, but this was a nice report. This actually 30% comes from England the UK. Also, some of the 30% 25% Azoles isolates that were submitted to this came from 25% Azoles evaluated: Itraconazole 20% 20% Voriconazole 20% the Netherlands. In this study, over the years, 20% Posaconazole 17% 17% 15% 14% 15% 14% they looked at the rate of resistance to azoles in 10% 10% aspergillosis. So, this was a reference lab. And 7% 7% 7% 7% 5% 5% 5% 5% 5% 5% 5% 5% 3% you can see there was an increase over time. 3% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% They specifically looked at itraconazole, voriconazole and posaconazole, because those Ahmed-Bueid et al. J Antimicrob Chemother. 2010;65: Ahmed-Bueid et al. J Antimicrob Chemother. 2010;65: were the agents that were available. So, there seems to be some increase that has been reported, and this was just one of many studies that are indicating that. Of course, the rates are still not very high, but the concern is that they re increasing. Annual Rates of of Mucormycosis-Associated Hospitalization Per Per 1 Million Persons: United States, Average annual annual percent change: +5.15%; p< The last thing I want to point out is that now we do have some drugs that are available for some mucormycosis indications. It turns out that actually Mucor is one of the nastiest fungal diseases that we see, and there has been some increase reported in the literature with mucormycosis. Some of this may also reflect diagnostics and factors like the increased use of steroids. Vallabhaneni et al. In: Proceedings from IDWeek 2016; October 26-30, 2016; New Orleans, LA. Abstract Vallabhaneni et al. In: Proceedings from IDWeek 2016; October 26-30, 2016; New Orleans, LA. Abstract Estimated Trends in in Antifungal Use: United States, Year Year All All Antifungals Antifungals Azole Azole Echinocandin Polyene Polyene Vallabhaneni et al. In: Proceedings from IDWeek 2016; October 26-30, 2016; New Orleans, LA. Abstract Vallabhaneni et al. In: Proceedings from IDWeek 2016; October 26-30, 2016; New Orleans, LA. Abstract The last slide I m going to show before we get to our next speaker is from the CDC, in which they collected some data looking at antifungal use in US hospitals over the past several years. You can see this is the answer to our one question. The class that has consistently shown an increase over that time has been the echinocandins. There was a time we were using a lot more azoles. We were using more azoles in our critical care units, and I think we ve trimmed some of that back. Especially, moving forward with the new guidelines from the IDSA, we may even continue to see certain azoles decrease in that Candida setting. 10

12 So, that s all I have. And at this point I d like to welcome our next speaker, Dr. Ryan Shields, so please join me in welcoming him. Blood cultures remain negative in in approximately what percentage of of invasive candidiasis cases? Audience Response We re going to get started and the topic of this presentation is Diagnostic Testing - The Need for Speed. We re going to start this session with another audience response question. This questions asks: In what percentage of blood cultures do they remain negative despite patients actually having invasive candidiasis? So, blood cultures remain negative in approximately what percent of invasive candidiasis cases? The correct answer is actually 50%, which may surprise some people. So, up to 50% of patients with invasive candidiasis actually have blood culture-negative disease, and we ll talk about some of the reasons why that is. Spectrum of Invasive Fungal Infections (IFI) 4% 4% 3% 3% 2% 2% Candida 6% 6% Aspergillus 8% 8% 42% Other mold Mucor Cryptococcus 29% Endemic PCP I want to kick off where Dr. Slain left off. And really, what I m going to talk about, and what we re going to focus on for the rest of the presentation today, is three of the more common causes of invasive fungal infections. Those are Candida, Aspergillus and Mucor. Pfaller MA, et al. Clin Infect Dis. 2006;43:S3-S14. Pfaller MA, et al. Clin Infect Dis. 2006;43:S3-S14. Vazquez JA, et al. Ther Adv Infect Dis. 2013;1: Vazquez JA, et al. Ther Adv Infect Dis. 2013;1:

13 Signs and Symptoms of IFI Invasive Invasive Aspergillosis candidiasis Aspergillosis candidiasis Hematologic Hematologic Any Any hospitalized hospitalized malignancy malignancy Mucormycosis Mucormycosis Hematologic Hematologic malignancy malignancy When you look across the diseases that these fungal pathogens cause, there are some similarities and some important differences. Patients Patients patient patient with with risk risk Bone-marrow Bone-marrow Bone-marrow Bone-marrow factors We ll start with invasive candidiasis, which is factors transplant transplant transplant transplant Solid-organ Solid-organ transplant transplant Patients Patients with with diabetes diabetes really a disease that can affect any hospitalized Pulmonary Pulmonary Rhino-orbital-cerebral Candidemia Clinical Candidemia Clinical aspergillosis aspergillosis mucormycosis Deep-seated mucormycosis manifestations Deep-seated Tracheobronchitis Pulmonary patient. This is the common fungal infection that candidiasis Pulmonary candidiasis Rhinosinusitis Rhinosinusitis mucormycosis mucormycosis really brings all of our hospitals together, because Fever Fever Fever Wide Widerange Fever Pleuritic Pleuritic chest chest pain pain Nasal Nasal ulceration Signs Signs and and from from low-grade ulceration low-grade in general most all of us see invasive candidiasis. Hemoptysis Hemoptysis Periorbital Periorbital swelling symptoms swelling symptoms fever fever to to septic septic Shortness Shortness of of breath breath Decreased Decreased vision shock vision shock Cough Cough Headache Headache This disease typically manifests in two different Roden MM, et al. Clin Infect Dis. 2005;41: Roden MM, et al. Clin Infect Dis. 2005;41: Segal BH. N Engl J Med. 2009;360: diseases, either candidemia, a bloodstream Segal BH. N Engl J Med. 2009;360: infection, or deep-seated candidiasis, typically in the intra-abdominal space. And the symptoms range widely, from typically asymptomatic patients who may have low-grade fevers, to the other end of the spectrum that patients develop septic shock and multi-organ failure. Aspergillosis and mucormycosis, of course, are mold infections and have different clinical manifestations. These are typically opportunistic fungal pathogens that affect immunosuppressed patients, specifically patients with hematologic malignancies or those undergoing bone marrow transplantation or other types of solid-organ transplant. For aspergillosis, we predominantly see pulmonary disease, so pulmonary aspergillosis. We can also see tracheobronchitis and, more rarely, rhinosinusitis, and the symptoms are consistent with diseases we see. Typically, patients have fevers, pleuritic chest pain, hemoptysis, shortness of breath and cough, and the first three symptoms there are kind of the classic triad for identifying invasive candidiasis, particularly among neutropenic patients. The difference with Mucor is that this causes more rhino-orbital disease, particularly in the sinuses, although it can cause pulmonary disease as well; and the symptoms are, again, consistent with that. We see nasal ulcerations, periorbital swelling, decreased vision and headaches. Diagnosis of IFI Invasive Invasive candidiasis candidiasis Aspergillosis Mucormycosis Culture Culturesensitivity % 80% (Blood) (Blood) Up Upto to 50% 50% (BAL) Up (BAL) Upto to 50% 50% (BAL) 20 (BAL) 20 40% 40% (DSC) (DSC) ~70% ~70% (tissue) (tissue) Time Timeto to culture culture positivity positivity days days (up (up to to 88 days) days) days days (up (up to to days) days) days days (up (up to to days) days) Broad, Broad, ribbonlike, Yeast ribbonlike, Yeast cells cells µm µm Narrow, Narrow, septated Microscopic septated Microscopic nonseptated nonseptated hyphae (much hyphae (much larger larger than than hyphae hyphae with with acuteanglangle branching branching branching branching evaluation acute- evaluation with with right-angle bacteria) right-angle bacteria) Histopathology DSC DSConly Critical Critical Critical Critical Imaging Imaging DSC DSC only only Critical Critical Critical Critical Delayed Delayed treatment treatment associated associated with with worse worse outcome? outcome? Yes Yes Yes Yes Yes Yes Clancy CJ, et al. Clin Infect Dis. 2013:56: ; Segal BH. N Engl J Med. 2009;360: ; Horvath JA, et al. Am J Med. Clancy CJ, et al. Clin Infect Dis. 2013:56: ; Segal BH. N Engl J Med. 2009;360: ; Horvath JA, et al. Am J Med. 1996;100: ; Badiee P, et al. Iran J Microbiol. 2013;5: ; Walsh TJ, et al. Clin Infect Dis. 2012;54:S ;100: ; Badiee P, et al. Iran J Microbiol. 2013;5: ; 24 Walsh TJ, et al. Clin Infect Dis. 2012;54:S Despite the differences in these three diseases, in diagnosing them there are three really important points that bridge these diseases together. Point number one is that routine cultures are fairly insensitive for diagnosing all of these diseases. You see the percentages listed across there, but in general, our routine cultures that we re sending for our patients to the microbiology lab only pick up about 50% of these invasive fungal infections. So, number one: routine cultures are insensitive. Number two: routine cultures take time. This time results in a delay in treatment. The typical time to culture positivity is a number of days two to three days for candidiasis, and longer for the mold infections. And we know that if cultures do turn positive, they typically turn positive late in the disease course, and that has a direct correlation then for delayed treatment. 12

14 The other thing that we see, and point number three, is that we know for all of these invasive fungal infections, delayed treatment is associated with worse patient outcomes. Are there, then, opportunities that we as pharmacists have to help improve this delayed treatment and start antifungals sooner, and thereby improve patient outcomes? I think this is really an important avenue for pharmacists to be involved with. Invasive Candidiasis: What Are We Trying to Diagnose? Low Low incidence incidence among among large large atrisrisk populations populations at- (Medical) (Medical) Candidemia DSC DSC w/ w/ candidemia Leroy O, et al. Crit Care Med. 2009;37:1612; Clancy CJ, et al. Clin Infect Dis. 2013;56:1284. Leroy O, et al. Crit Care Med. 2009;37:1612; Clancy CJ, et al. Clin Infect Dis. 2013;56: Deep-seated candidiasis (DSC) (DSC) 39% 31% 28% High High incidence incidence among among narrow narrow atrisrisk populations populations at- (Surgical) (Surgical) Sensitivity Missing of of blood blood ~50% ~50% of of cultures % % disease We ll start with invasive candidiasis. With this disease, it s important to know what exactly we re trying to diagnose. Invasive candidiasis really encompasses three entities. One is candidemia, which is a low-incidence disease among wide populations. This includes all hospitalized populations and predominantly our medical patients, and we see that the sensitivity of blood cultures for candidemia is about 60 to 80%. So, in patients that have true bloodstream infections the sensitivity for blood cultures is pretty good. On the other end of the spectrum is deep-seated candidiasis, i.e., intra-abdominal candidiasis and other deep-seated diseases, which is typically a high-incidence disease among very narrow patient populations. So, these are your abdominal surgery patients that the surgeon goes in and tells you they find this nice abscess. And of course, they didn t send it for culture, but we know a lot of these abscesses are consistent with Candida infections. In this case, blood cultures are much less sensitive, because we know that Candida from deep-seated sites only enters the bloodstream in a minority of cases, and that s in the middle of the Venn diagram here, and that s deep-seated candidiasis with candidemia. If you look cumulatively then, and this is the answer to our initial audience response question, we re missing about 50% of the disease across the spectrum of invasive candidiasis. 13

15 Microbiological Evidence Definitions of Early Antifungal Treatment Proven Proven Late Late Targeted Targeted disease disease antifungals antifungals Empiric Empiric Preemptive Preemptive Prophylaxis Prophylaxis Disease Disease unlikely unlikely Early Early antifungals antifungals Likelihood Likelihood of of Disease Disease Adapted from: Eggimann et al. Ann Intens Care. 2011;1:37. Adapted from: Eggimann et al. Ann Intens Care. 2011;1: A lot of terms get thrown around in initiating early antifungal therapy, and I want to put us all on the same page in how we re going to use these terms today. Prophylaxis, as Dr. Slain had pointed out, is really the idea of using antifungals in the absence of disease among high-risk populations. So, you see, for prophylaxis, the likelihood of the disease is the lowest and you have the least amount of microbiologic evidence, but we know that these are patients at high risk for invasive candidiasis, so we prophylaxis them against fungal infections. Preemptive therapy is again using antifungals before the onset of signs and symptoms. That s an important part that we re going to talk about today the idea of using non-culture-based diagnostic tests to find the right patient populations in which we can give them preemptive antifungals before the onset of signs and symptoms of the disease. Empiric therapy is at the onset of signs and symptoms of disease. So, this would be your patient that has had fevers in the ICU for four days despite being on broad-spectrum antibiotics. Initiating early empiric antifungal therapy may help us get early antifungal therapy on board. And then finally, targeted therapy is at the point where we have a positive culture and we re initiating antifungal therapy in response to that positive culture. So, this is the highest amount of microbiologic evidence, but also you re initiating antifungal therapy at a delayed point, after the organism s already been identified. This is a busy slide, but I want to give you a Invasive Candidiasis Across Populations couple of takeaways for invasive candidiasis here. Risk Risk of of IC* IC* Patient Patient examples examples Type Type of of IC IC Incidence Incidence Any hospitalized patient w/ a blood culture Because we know that the incidence of invasive Any hospitalized patient w/ a blood culture Low Residence in the ICU without further risk stratification Low Residence in the ICU without further risk stratification Candidemia Candidemia <1% <1% Residence in the ICU post-cardiothoracic surgery Residence in the ICU post-cardiothoracic surgery Low Low incidence incidence candidiasis ranges very widely across patient Peritoneal dialysis with peritonitis Deep-seated Peritoneal dialysis with peritonitis Deep-seated 3 3 6% Low-tomoderate Uncomplicated liver transplant recipient Low-tomoderate Presence of septic shock Uncomplicated liver transplant recipient candidiasis 6% candidiasis populations, this is important for how we re Presence of septic shock Candidemia Candidemia 3 3 7% 7% Empiric Empiric vs ICU residence for 4 days vs ICU residence for 4 days Preemptive Preemptive going to initiate early antifungal therapy. So, we Deep-seated Liver transplant recipient + other risk factors for IC Deep-seated Liver transplant recipient + other risk factors for IC % candidiasis 20% Moderate candidiasis Moderate know at the high end of the incidence, we have ICU residence for 4 days + other risk factors for IC ICU residence for 4 days + other risk factors for IC Candidemia Candidemia % 15% Severe acute or necrotizing pancreatitis Prophylaxis Severe acute or necrotizing pancreatitis Prophylaxis high-risk patients. These are liver transplant Liver transplant recipient with bile leak + other risk Deep-seated High Liver transplant recipient with bile leak + other risk Deep-seated High % factors for IC 40% factors for IC candidiasis candidiasis Recurrent GI track leak requiring surgery Recurrent GI track leak requiring surgery patients with bile leaks, for instance, or other Prophylaxis Prophylaxis is is beneficial beneficial in in preventing preventing invasive invasive fungal fungal infections infections among among populations populations with with baseline baseline rates rates of of disease disease 15% 15% abdominal surgery patients that have ongoing Clancy CJ, Shields RK, Nguyen MH. J Fungi. 2016;2:10. Clancy CJ, Shields RK, Nguyen MH. J Fungi. 2016;2: surgeries. Their incidence of invasive candidiasis at the beginning is somewhere between 20 to 40%. And if you look across the literature, we know that when you have a patient population with an incidence of disease that exceeds 15%, prophylaxing them with antifungal therapy shows benefit. We use this benchmark as 15% of really where we want to initiate antifungal therapy, when we have an incidence of at least 15%. We know our high-risk patients meet that, and these are patients we re typically going to use prophylactic strategies for. 14

16 On the other end of the spectrum, we have patients that are at risk for invasive candidiasis, but the incidence is so low. These are hospitalized patients with any blood culture that s sent, or maybe your low-risk ICU patient, such as those in the cardiothoracic ICU, where the incidence of invasive candidiasis is so low that using early antifungal strategies and trying to identify those patients is tough, because you have to screen so many patients to find those one or two that might be at risk for the disease. But in the middle, there s this sweet spot where the incidence of invasive candidiasis ranges from anywhere from 3 to 15%, and this is the place where empiric and preemptive antifungal strategies really might be important for our patients moving forward. What Are Are the the Non-culture Diagnostic Options for for Candida? Pooled Pooled Pooled Pooled Test Test Sensitivity Sensitivity Specificity Specificity Comments Comments and and Limitations Limitations Combined Combined test test is is superior superior (IgG (IgG > > IgM) IgM) Antigen/antibody Positive Positive before before blood blood cultures cultures in in % 73% (mean (mean of of days days prior) detection prior) detection 83% 83% 86% 86% Limitations: (Mannan/ Limitations: (Mannan/ antimannan) antimannan) Antigens Antigens may may be be rapidly rapidly cleared cleared from from blood blood Concerns Concerns with with immunosuppression immunosuppression Only Only prospective prospective study study showed showed low low specificity specificity FDA FDA approved approved as as an an adjunct adjunct to to culture culture β-d-glucan β-d-glucan ~75% ~75% Limitations: 56 Limitations: 56 93% 93% False False positives positives (P. (P. aeruginosa, aeruginosa, albumin, albumin, HD) HD) Less Less sensitive sensitive for for deep-seated deep-seated candidiasis candidiasis Used Used to to initiate initiate early early antifungals antifungals in in sepsis sepsis Positive Positive before before blood blood cultures cultures (1 ( days) Nucleic days) Nucleicacid acid 95% 95% 92% 92% Prognostic Prognostic value value if if persistently persistently positive detection positive detection (PCR) (PCR) Limitations: Limitations: Lack Lack of of standardization standardization across across assays assays Clancy CJ, et al. Clin Infect Dis. 2013;56: ; Mikulska M, et al. Crit Care. 2010;14:R222; Clancy CJ, et al. Clin Infect Dis. 2013;56: ; Mikulska M, et al. Crit Care. 2010;14:R222; Karageorgopoulos DE, et al. Clin Infect Dis. 2011;52:750-70; Avni T, et 28al. J Clin Microbiol. 2011;49: Karageorgopoulos DE, et al. Clin Infect Dis. 2011;52:750-70; Avni T, et 28al. J Clin Microbiol. 2011;49: Let me show you how we re going to incorporate some of these things. First, let s talk about some of the tests that are available. These are the nonculture-based diagnostic options for Candida, the first of which is an antigen antibody detection method. This detects mannan antigens and antimannan antibodies. This is a test that s been around for a long time, and meta-analyses data have shown us that the pooled sensitivity and specificity is about 85% for these tests, meaning that they are fairly good tests. But there are some problems with antigen and antibody-mediated tests, first of which is, antigens are rapidly cleared from the blood. Mannan antigens mannan is a part of the fungal Candida cell wall is rapidly cleared from the blood, so it s hard to detect early. The other problem is that there are concerns with immunosuppression. If you have patients that are on immunosuppressive agents, they may not mount the same sort of antibody responses as an otherwise immune-competent host would. Because of that, there are very few prospective studies that have looked at mannan and anti-mannan detection methods, and those have shown a relatively lower specificity, so in general these tests aren t used clinically all that commonly. The second test we ll talk about is detection of beta-d-glucan. Again, beta-d-glucan is a major constituent of the Candida cell wall, and this is an FDA-approved test as an adjunct to culture. Here, your pooled sensitivity is about 75%, but you see the specificity ranges widely. The reason is there is the potential for false positive tests here. We know that certain gram-negative pathogens like Pseudomonas can elicit a false positive glucan assay, as well as human blood products like albumin or IVIG. And then, we know that glucans can be found in cellulose filters so, filters that are used in hemodialysis or other types of membranes also contain glucans. We know that these things can cause false positives. The other thing that is important about beta- D-glucan is it s less sensitive for deep-seated candidiasis, which we know is a very important entity for invasive candidiasis overall. The third test we ll talk about for early diagnosis of Candida is a nucleic acid detection, or a PCRbased assay. This detects Candida DNA using PCR probes, and this can be used to initiate early 15

17 antifungal therapy. It s typically positive before blood cultures are, and it has persistently positive prognostic value, which we ll talk about. The major problem with PCR assays is, despite having a very high sensitivity and specificity, there s no standardization across these assays. There are a lot of mom-and-pop shops out there that have their PCR-based assays for Candida, but between them there s no standardization. They all use different primers and different probes to detect Candida, and there are no FDA-approved PCR-based tests at this point. There are companies working on this, and hopefully in the future we can have some standard PCRbased assays, because you can see the sensitivity and specificity for these assays is really good. Most recently, however, there s a new Newest Technology technology that has entered this space for nonculture-based-type diagnostics, and that s a Magnetic resonance technology (T2Candida) magnetic resonance technology. T2Candida is FDA approved as as an adjunct to to cultures Results from whole blood samples in in ~4 hours the first company to incorporate this technology Identifies the 5 major Candida species into their platform. This is also an FDA-approved Limitations: assay that uses these nanoparticles coated with Lack of of clinical data (but this is is improving!) Candida-binding components that, in whole Not studied in in deep-seated Candidiasis Is Is it it too sensitive? blood samples, bind Candida, and uses the same technology as MRIs to detect Candida at a very Mylonakis E, et al. CID. 2015;60:892. Mylonakis E, et al. CID. 2015;60: high sensitivity and specificity. Here, with this assay, we can get results from whole blood samples in four hours. So, let s compare that to what we re doing with routine blood cultures for diagnosing candidemia, which typically turn positive about three or four days after we collect the blood culture; then it may take another day or two to identify the organism, and then potentially another day to do susceptibility testing. Here, you re identifying the bug in four hours. So, it s a huge opportunity to initiate antifungal therapy sooner. It also uses a multiplex PCR-based platform to identify the five major Candida species that Dr. Slain pointed out. But there are limitations to this new technology. There are not a lot of clinical data yet, and we re certainly not sure how to best incorporate this into clinical practice. Also, it hasn t been studied for deep-seated candidiasis. The final limitation is that we wonder if some of these assays may become too sensitive. What do you do when your non-culture-based diagnostic test is positive but your blood culture is negative? Does that mean the patient really has a blood culture or a bloodstream infection? We don t know the answers to some of these questions yet. Let me pause there for a second and ask the audience a question. How many of you in your hospitals are using some of these non-culture-based diagnostic methods to detect Candida in your patients? By a show of hands, anybody out there? It looks like in the vast minority of patients. 16

18 So What s the Problem? Improved technology Non-culture-based diagnostics are highly sensitive and specific Able to to identify major Candida species FDA approved So, what s the problem? We certainly have an improved technology and a number of nonculture-based diagnostic assays that are now FDA-approved and that can even identify Candida to the species level. We have the technology and we have the test available. Why aren t we all using them? Well, one reason why is that these tests are So What s the Problem? expensive. It costs money to order tests several Tests are expensive hundreds of dollars in some cases and we re Compared with suboptimal gold comparing them to a suboptimal gold standard. standard Takes time to to redefine Remember, blood cultures are only sensitive for treatment approaches about picking up 50% of the disease. So, we Which patients will benefit? don t know the true burden of disease, because we re comparing to a suboptimal gold standard. And we know with new technology it takes time to refine some of these approaches. I want to help you try to refine some of these approaches for your patients. We ll talk next about what patients are likely to benefit. Invasive Candidiasis Across Populations Risk Risk of of IC* IC* Patient Patient characteristics characteristics Type Type of of IC IC Incidence Incidence PPV* PPV* NPV* NPV* Any hospitalized patient w/ a blood culture Any hospitalized patient w/ a blood culture Residence in the ICU without further risk Low Residence in the ICU without further risk Low stratification stratification Candidemia Candidemia <1% <1% 8% 8% >99% >99% Residence in the ICU post-cardiothoracic surgery Residence in the ICU post-cardiothoracic surgery Low-tomoderate Low-tomoderate Moderate Moderate High High Peritoneal dialysis with peritonitis Peritoneal dialysis with peritonitis Uncomplicated liver transplant recipient Uncomplicated liver transplant recipient Presence of septic shock Presence of septic shock ICU residence for 4 days ICU residence for 4 days Liver transplant recipient + other risk factors for IC Liver transplant recipient + other risk factors for IC Deep-seated Deep-seated candidiasis candidiasis 3 3 6% 6% % 14% 99% 99% Candidemia Candidemia 3 3 7% 7% % 40% 99% 99% Deep-seated Deep-seated % 20% 12 candidiasis 12 40% 40% 98% 98% candidiasis ICU residence for 4 days + other risk factors for IC Candidemia 10 15% 50 61% 98% ICU residence for 4 days + other risk factors for IC Candidemia 10 15% 50 61% 98% Severe acute or necrotizing pancreatitis Severe acute or necrotizing pancreatitis Liver transplant recipient with bile leak + other risk Liver transplant recipient with bile leak + other risk factors for IC factors for IC Recurrent GI track leak requiring surgery Recurrent GI track leak requiring surgery Deep-seated Deep-seated % 40% candidiasis 64% 64% % 93% candidiasis *Assumptions based based on on sensitivity/specificity of of 90%/90% 90%/90% and and 80%/70% 80%/70% for for I hope everyone s had their coffee this morning, because we re going to get nerdy here for just a minute; but bear with me, because I think these are really important data. So, if we take the same slide as before and we look at the incidence of invasive candidiasis in the same patient populations, look at the positive and negative predictive values of some of these nonculture-based diagnostics. candidemia and and deep-seated candidiasis, respectively. 32 By positive predictive value, we mean if you 32 have a positive non-culture-based diagnostic, a positive PCR assay, what s the likelihood that that patient has true invasive candidiasis? And on the other hand, negative predictive value is, if your test is negative, what are the odds that the patient does not have invasive candidiasis? So, we know in our highest-risk patients that already have a high incidence of disease, your positive test only helps you a little bit. It pushes your percentages up to give you some more 17

19 certainty, but it also doesn t have a great negative predictive value. Meaning, if your test is negative, there s still a high chance, maybe of 10 to 15%, that the patient truly has disease. So, for high-risk patients, these non-culture-based diagnostics may not be that beneficial. And if we use our cutoff that we talked about before as 15%, when we draw the positive predictive values here, we know that for our lowest-risk patients again, they re not hitting that threshold to initiate antifungal therapies of at least having a baseline incidence of 15%. So, we know that our lowest-risk patients may not benefit as well. But for these patients in the middle, where you see the positive predictive values now go from a baseline incidence of 3 to 15%, if you have a positive test, now these patients have a 21 to 60% chance of having disease, so that certainly helps you diagnostically. But perhaps the best utility of these agents is to look at these negative predictive values here in this sweet spot that we talked about. Now you have a 98% to 99% certainty that your patient does not have the disease. This could be used to incorporate strategies to stop antifungal therapy, or at least to withhold antifungal therapy, in patients that you have a high degree of certainty do not have the disease. So, the negative predictive value here is extremely important. Fagan Nomogram Applied to to Non-culture Diagnostics (+) (+) and and (-) (-) likelihood likelihood ratios ratios for for sensitivity/specificity of of 90% 90% are are 99 and and 0.11, 0.11, respectively Fagan TJ. N Engl J Med. 1975;293:257. Fagan TJ. N Engl J Med. 1975;293: Keeping with our nerdy theme for a second here, I want to point to an article that was published by Terry Fagan in the New England Journal of Medicine, where he used some Bayesian reasoning or Bayes Theorem to essentially establish this nomogram that says, if you know the pre-test likelihood of your patient having disease, you can use this formula using a likelihood ratio, and that and your non-culturebased diagnostic will give you some postprobability test of disease. If we use an example of a patient that has a baseline incidence of 10% for invasive candidiasis, and we use a sensitivity and specificity of 90%, which we know is true for PCR if you have a positive test, we know that the likelihood ratio is 9, and that they have a 50% chance of having invasive candidiasis. And similarly, if you have a negative PCR assay, they have a very low incidence of having invasive candidiasis. So, if you can model some of this, you can see how these tests help you across patient populations. 18

20 Sensitivity for for diagnosis Post-test Post-test probability probability of of invasive invasive candidiasis candidiasis Post-test Post-test probability probability of of candidemia candidemia Pre- and Post-test Likelihood for PCR Again, going with our example patient that has a baseline incidence of invasive candidiasis of 10%, in the absence of any of these culture diagnostics, they continue on this linear scale, where their pre-test and post-test likelihood of disease is the same. You know the baseline 0.50 Positive PCR Positive PCR Negative PCR Negative PCR (+) test = 50% probability of disease incidence, and that s all you know. But if you (+) test = 50% probability of disease No test 0.30 No test have a positive assay, like a positive PCR that has (-) test = 1.2% probability of disease (-) test = 1.2% probability of disease 0.00 a 90% sensitivity and specificity, what that assay Pre-test Pre-test probability probability of of candidemia candidemia tells you, then, if it s positive, is that your patient 34 now has a 50% probability of having disease; and 34 if your assay is negative, they only have about a 1% chance of disease. Ideal Ideal non-culture non-culture assay assay Absence Absence of of non-culture non-culture tests 0.90 tests High-risk for DSC High-risk for DSC Moderate risk for candidemia Moderate risk for candidemia Moderate risk for DSC 0.30 Moderate risk for DSC 0.30 Low-to-moderate risk for candidemia Low-to-moderate risk for candidemia 0.20 Low-to-moderate risk for DSC 0.20 Low-to-moderate risk for DSC High-risk for DSC High-risk for DSC Proposed cutoff to initiate antifungals 0.10 Proposed cutoff to initiate antifungals 0.10 Low risk Low risk Moderate risk for candidemia Moderate risk for candidemia Moderate risk for DSC Low-to-moderate Moderate risk risk for for candidemia DSC 0.00 Low-to-moderate risk for candidemia Low-to-moderate risk for DSC 0.00 Low-to-moderate risk for DSC Pre-test Pre-test probability probability of of invasive invasive candidiasis candidiasis these assays are particularly important. So, what these non-culture-based diagnostic tests really can do, then, is shift the paradigm, where if we use our threshold of 15% as our cutoff to initiate antifungal therapy in the absence of non-culture-based diagnostics, we re really only initiating antifungal therapy in our highest-risk patients. But with non-culturebased diagnostics, what you can do is initiate early antifungal therapy in a much wider variety of patient populations, including patients with low to moderate risk of invasive candidiasis. So, I want to transition into talking a little bit about Diagnosing Aspergillosis diagnosing invasive aspergillosis. This is a moving target, right? Because we know for Non-culture diagnosing invasive aspergillosis, some of the Diagnostics Radiographic definitions are based on our degree of certainty Imaging Imaging that the patient has the disease either Biopsy Biopsy and and Histopathology possible, proven, or probable disease. Where I Culture Culture and and Microscopic Evaluation see non-culture-based diagnostics useful for Patient Patient Signs Signs and and Symptoms Symptoms Aspergillus is that it helps in our degree of certainty that the patient really has disease; but they have to be used in combination with all the other things we use to diagnose invasive aspergillosis. First of which is, you have to identify the right patients that have signs and symptoms of the disease. We use cultures, histopathology and radiographic imaging to help us as well. 19

21 Perhaps the most commonly-used culture aid, or Galactomannan (GM) Antigen Detection non-culture-based diagnostic, is galactomannan, Major constituent of of Aspergillus cell wall specifically, the galactomannan antigen Platelia assay FDA FDA approved for for serum and and BAL BAL samples Optical density (OD) is is a positive result detection. Galactomannan is another major Meta-analyses show sensitivity/specificity for for diagnosis: Serum: 71%/89% and and BAL: BAL: 87%/89% constituent of the Aspergillus cell wall, and this is Performs best among hematologic malignancy and stem cell transplant an FDA-approved test for both serum and BAL Less Less useful among non-neutropenic and and solid-organ transplant Decreases sensitivity by by administration of of antifungals samples, and it s measured on an optical densitybased report, where a value greater than 0.5 is Piperacillin/tazobactam, GI GI mucositis, blood products, other fungi positive. Meta-analyses show galactomannan is Caution for for false positive results! Miceli MH, Maertens J. Semin Respir Crit Care Med. 2015;36:650-61; Patterson TF, et al. Clin Infect Dis. Miceli MH, Maertens J. Semin Respir Crit Care Med. 2015;36:650-61; Patterson TF, et al. Clin Infect Dis. fairly sensitive and specific, and this test really 2016;63(4):e1-e60; Pfeiffer CD, et al. Clin Infect Dis. 2006;42: ;63(4):e1-e60; Pfeiffer CD, et al. Clin Infect Dis. 2006;42: performs best in your hematologic malignancy and stem cell transplant patients. It s less useful for non-neutropenic patients, specifically solidorgan transplant patients, and it s unclear if that s because the disease is different in these patients, or they just haven t been studied to the same degree. It also has decreased sensitivity with the initiation of antifungals. Now, galactomannan can also give you false positive results, which is very important. One of the things for years and years that has been talked about is the false positive results with piperacillin/tazobactam, and the formulation of that drug. That s become less of an issue most recently. The manufacturer really hasn t released why, but the formulation now does not appear to have cross-reactivity with galactomannan, so I see that as less problematic moving forward. Because of that, there are fewer false positive galactomannan results. Aspergillus Non-culture Diagnostics Test Test Pooled Pooled Sensitivity Sensitivity Pooled Pooled Specificity Specificity Comments Comments and and Limitations Limitations β-d-glucan β-d-glucan 77% 77% 85% 85% FDA FDA approved approved in in serum serum (Fungitell) (Fungitell) >80 >80 pg/ml pg/ml = = positive positive Limitations: Limitations: Cannot Cannot distinguish distinguish fungal fungal species species False False positives positives (same (same as as Candida) Candida) High High sensitivity sensitivity for for Aspergillus Aspergillus DNA DNA May May detect detect Aspergillus Aspergillus before before GM 84% GM 84% 76% Nucleic 76% Nucleicacid acid Two Two positive positive results results increases increases specificity (Serum) specificity (Serum) (Serum) detection (Serum) detection Limitations: 77% Limitations: 77% 94% (PCR) 94% (PCR) Unable Unable to to differentiate differentiate colonization (BAL) colonization (BAL) (BAL) (BAL) DNA DNA released released sporadically sporadically during during infection infection No No assay assay standardized standardized and and validated validated Detects Detects mannoprotein mannoprotein with with monoclonal monoclonal Ab 20 Ab 20 68% 68% % 98% Point Point of of care care test Lateral test Lateralflow flow (Serum) (Serum) (Serum) (Serum) device Inexpensive, Inexpensive, no no equipment equipment required required device (LFD) (LFD) % 100% % 95% Limitations: (BAL) Limitations: (BAL) (BAL) (BAL) Performance Performance varies varies across across populations populations Your other option for diagnosing Aspergillus early is beta-glucan. Just like Candida, beta-glucan is also FDA-approved and has a positive value greater than 80, but it cannot distinguish between fungal species. So, if you have somebody with both Candida and aspergillosis, this assay will turn positive, but it s not going to tell you which fungal pathogen you re trying to diagnose. Miceli MH, Maertens J. Semin Respir Crit Care Med. 2015;36:650-61; Arvanitis M, et al. J Clin Microbiol. 2014;52:3731; Miceli MH, Maertens J. Semin Respir Crit Care Med. 2015;36:650-61; Arvanitis M, et al. J Clin Microbiol. 2014;52:3731; Hoenigl M, et al. J Clin Microbiol. 2014;52:2039. Hoenigl M, et al. J Clin Microbiol. 2014;52: There are also PCR-based assays for Aspergillus, which also are fairly sensitive and specific. We know that, with all these assays, if you have at least two positive results, that increases your specificity, and that s certainly true for PCR as well. PCR, keep in mind, identifies DNA of the fungal pathogen, so it s not going to tell you the difference between colonization versus infection. You ll need some clinical insight to do that. Finally, lateral flow devices are a relatively new player in this field. This is an inexpensive, rapid diagnostic test with no equipment required that detects a mannoprotein with monoclonal antibodies, and really can be used as a point-of-care test. These lateral flow devices can give you results in as quickly as ten minutes, and that s particularly useful for even doing at the patient s bedside from a BAL sample. 20

22 There s been one study that s looked at all these Comparison of Diagnostic Methods assays together and given us some indication of 78 patients what s the best assay for identifying aspergillosis All All tests performed in in patients (10 with IA) IA) Sensitivity Sensitivity Specificity Specificity in your patient. And what you can see from the 100% 95% 98% 100% 100% 100% 95% 98% 100% 95% 95% 95% 98% 100% 95% 98% 90% 90% 90% 90% 90% 93% 90% 93% slide here is that we know cultures are relatively 80% 80% 80% 80% 80% 80% 76% 80% 76% 80% 76% 70% 76% 70% poor again, keeping consistent with our 60% 60% 50% 50% 40% theme, 50% sensitivity and you see that, 40% 20% 20% individually, galactomannan, beta-d-glucan, PCR 0% 0% and lateral flow devices do okay. But you can Culture Culture GM GM BDG BDG PCR PCR LFD LFD BDG BDG +/- +/-LFD +/- +/- GM GM +/- +/- GM GM +/- +/- PCR PCR PCR PCR PCR PCR LFD LFD have some benefit if you combine tests, and if Hoenigl M, et al. J Clin Microbiol. 2014;52:2039. Hoenigl M, et al. J Clin Microbiol. 2014;52: you look specifically at the idea of having either a positive galactomannan or positive PCR, you now have 100% sensitivity and 98% specificity. So, perhaps using these tests in combination is a way forward for the field. What s the Role of Diagnostics for IA? Using tests in in combination improves sensitivity Best Best used in in the the context of of high high suspicion for for Aspergillosis Arvanitis Arvanitis M, M, et et al. al. Clin Clin Infect Infect Dis. Dis. 2015;61:1263 Screening with GM and PCR among hematologic malignancy and HSCT results in: in: Reduced use use of of antifungal agents Morrissey Morrissey CO, CO, et et al. al. Lancet Lancet Infect Infect Dis. Dis. 2013;13:519 Earlier diagnosis of of IA IA Aguado Aguado JM, JM, et et al. al. Clin Clin Infect Infect Dis. Dis. 2015;60:405 But not not validated in in randomized trials Negative predictive value is is extremely useful! Here s my take on the role of diagnostics for invasive aspergillosis using the tests in combination improves sensitivity, particularly repeating the test. If you have a negative PCR, and you repeat that, having two negatives gives you almost 100% certainty that you do not have the disease. Is Is there a role role in in antifungal stewardship? There have been screening strategies proposed Maertens Maertens J, J, et et al. al. Clin Clin Infect Infect Dis. Dis. 2005;41:1242 for both galactomannan and PCR, particularly among hematologic malignancy and stem cell transplant patients. I m less in favor of this. I think the more you look for positive tests, the more you can find them. There s a high probability that you find positive tests that you don t know how to incorporate. To me, the usefulness of these tests is in the right patient populations where you have a high clinical suspicion of disease. This is where you should be using these tests. And the negative predictive value, just like Candida, is extremely useful. The negative predictive value here is so high it essentially excludes the possibility of disease, at least at that time point. 21

23 Early Diagnosis of Mucormycosis BDG and GM are NOT useful! Mucor-specific PCR has been investigated, with promising results: Among patients with (+) (+) culture, PCR was (+) (+) in in (83%) Among patients with (-) (-) culture, PCR was (+) (+) in in (80%) Clinicians must rely on culture, histopathology, and clinical suspicion Briefly, then, my last slide on mucormycosis. Beta-D-glucan and galactomannan are not useful. These are not major parts of the Mucor cell wall. There have been PCR assays investigated that have shown promising results so far, but these things are a long way away from being primetime to this point. Hammond SP, et al. J Clin Microbiol. 2012;49:2151; Walsh TJ, et al. Clin Infect Dis. 2012;54:S55. Hammond SP, et al. J Clin Microbiol. 2012;49:2151; Walsh TJ, et al. Clin Infect Dis. 2012;54:S Summary Delayed antifungal therapy results in in worse patient outcomes across IFIs Routine cultures take time and are are poorly sensitive Non-culture based diagnostic tests may help to to improve approaches to to early initiation of of therapy Less commonly employed across invasive candidiasis than aspergillosis; however, new strategies are are being investigated Need to to identify the the right patient populations Negative predictive value may be be the the best use of of the the test populations for these tests to be useful In summary, what we can say is that we know delayed antifungal therapy is associated with worse patient outcomes, and routine cultures take time and are poorly sensitive. Because of that, non-culture-based diagnostics are really important for our ability to be able to improve our early initiation of antifungal therapy. These are less commonly employed for invasive candidiasis but used fairly commonly for aspergillosis. But the key for all these things is that you have to identify the right patient With that, I thank you for your attention, and we ll be happy to turn it over to my co-speaker today, Dr. Johnson. Thank you. Now we re going to turn our attention to talking about our approach to therapy, and choosing between the available antifungal agents for patients at risk of, or who are experiencing, an invasive fungal infection. 22

24 Which of these antifungals is NOT recommended as monotherapy for invasive pulmonary aspergillosis? 44 Audience Response First, an audience response question: Which of these antifungals is not recommended as monotherapy for invasive pulmonary aspergillosis? And the correct answer is C, caspofungin. As we ll discuss in a few moments, caspofungin has data in salvage therapy but not initial therapy, or at least not an FDA indication for initial therapy; and the other drugs listed here do have FDA indications as initial therapy for treatment of invasive pulmonary aspergillosis. So, when thinking about our approach to these Defining Risks: Yeast vs Mold Infections patients and how we manage them, we really Yeasts Molds do need to think about those risks. Both the Neutropenia Yes Yes Yes, Yes, Prolonged prior speakers did cover a little bit of this, and Mucositis Yes Yes No No we ll talk about it a little bit later again in T-cell T-cell immunodeficiency No No Yes Yes terms of how you approach empirical and Respiratory viral viral disease No No Yes Yes Environmental exposure No No Yes Yes prophylactic therapy for these patients. It Hyperglycemia Yes Yes Yes Yes largely depends on their risk. Are they more at Vascular catheter Yes Yes No No risk for yeast? Are they more at risk for mold? Or, are they at risk for both? And we know Nucci M, Anaissie E. Blood. 2014;124: Nucci M, Anaissie E. Blood. 2014;124: that the transmission mechanisms for these are a little bit different. Yeasts are a commensal organism. They live in the gut and on the skin for Candida. And so things like vascular catheters that allow a portal entry, or mucositis, which disrupt the GI tract, allow yeast to get into the bloodstream and set up shop and cause infection. But, for molds, we know that they re transmitted primarily by spores in the environment. So, environmental exposures are very important for these, as well as things like respiratory viral diseases, which affect the immune system locally in the lungs and allow then an environment in which the fungi can overgrow, and lead to infection. 23

25 Systemic Antifungal Agent Pipeline Posaconazole Posaconazole Anidulafungin Anidulafungin Micafungin Voriconazole Micafungin Voriconazole Caspofungin ABLC Caspofungin ABLC Itraconazole Itraconazole Terbinafine Terbinafine L-AmB L-AmB 5FC 5FC Amphotericin Amphotericin BB Fluconazole Fluconazole Adapted Adapted from from Ostrosky-Zeichner Ostrosky-Zeichner L. L. Nature Nature Reviews Reviews Drug Drug Discovery. Discovery. 2010;9: ;9: Looking at the available systemic antifungal agents, we now have at least 13 different antifungal agents that are available for systemic administration in the United States. We ve had sort of an explosion of antifungal development over the last 20 years. And the newest kids on the block here are isavuconazole as well as additional formulations of posaconazole. What I depicted here in green are also several new antifungal agents that are under development. So, the pipeline is not yet dry, and this is still an area that is ripe for more research. Looking at the available agents, they really do Comparative Clinical Trials: FDA Indications differ in terms of spectrum of activity and the AmB/ AmB/ clinical trials data that is used to support them. Infection LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa Invasive This is a summary slide where I show how these Candidiasis agents differ for each of the indications. And as Aspergillosis salvage salvage salvage salvage you can see in green, that indicates where we Mucormycosis* have clinical trials as well as an FDA indication for Prophylaxis use. And black just depicts where we have Neutropenic Fever Fever * * clinical trials data but no specific FDA indication. No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): And we ll use this as a roadmap as we talk about Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): these individual uses throughout the talk. But clearly, there s no one-size-fits-all approach to fungal infections in these patients. Pharmacokinetics Drug Drug T½ T½ (hr) (hr) Cmax Cmax (µg/ml) (µg/ml) Oral Oral Bioavailability Bioavailability (%) (%) CSF CSF penetration penetration (%) (%) Urine Urine (%) (%) Amphotericin Amphotericin deoxycholate deoxycholate <5 < Liposomal Liposomal amphotericin amphotericin B B <5 <5 <5 <5 5 5 ABLC ABLC <5 <5 <5 <5 <5 <5 Anidulafungin Anidulafungin <5 <5 <5 <5 <2 <2 Caspofungin Caspofungin <5 <5 <5 <5 <2 <2 Micafungin Micafungin <5 <5 <5 <5 <2 <2 Fluconazole Fluconazole >60 > Voriconazole Voriconazole <2 <2 Posaconazole Posaconazole OS OS OS; OS; DR DR (~24-31 (~24-31 IV, IV, DR) DR) (3.8 (3.8 IV, IV, DR) DR) (fasting) (fasting) ND ND <2 <2 Isavuconazole Isavuconazole % 98% + + preclin preclin?? Looking at the pharmacokinetics of the drugs, they do differ dramatically. In terms of oral bioavailability, we have limited oral bioavailability for both polyenes and echinocandins. However, I would mention that there are formulations of amphotericin B cochleates as well as an oral glucan synthase inhibitor that are under development and may change this landscape in the future. Dodds ES et al. Pharmacotherapy. 2000;20: Lewis RE. Current Fungal Infection Reports. 2008;2:5-11. Cresemba PI Dodds ES et al. Pharmacotherapy. 2000;20: Lewis RE. Current Fungal Infection Reports. 2008;2:5-11. Cresemba PI Lewis RE. Mayo Clin Proc. 2011;86: Kraft WK et al. Antimicrob Agents Chemother. 2014;58: Lewis RE. Mayo Clin Proc. 2011;86: Kraft WK et al. Antimicrob Agents Chemother. 2014;58: Maertens J et al. Antimicrob Agents Chemother. 2014;58: Aranda-Rajah MR, Kontoyiannis D. Future Med. Maertens J et al. Antimicrob Agents Chemother. 2014;58: Aranda-Rajah MR, Kontoyiannis D. Future Med. 2015;10: ;10: So, what we are left with, in terms of orally-available agents right now, are the azoles. These generally have at least 90% bioavailability and are well-tolerated when given that way. In terms of CSF penetration, we see dramatic differences as well. Fluconazole has the highest CSF penetration, while polyenes and echinocandins have very little. Despite this, polyenes seem 24

26 Trough Trough (ng/ml) (ng/ml) to be effective in treating things like cryptococcal meningitis and CSF infections because they have other immunomodulatory effects that make them effective in those settings. Urine continues to be a big challenge for these drugs as well. Fluconazole has the highest urinary excretion of all the antifungals here. We have much less urinary excretion for polyenes as well as echinocandins, and this creates a lot of challenges when dealing with it, because the urine can often be a source for many patients yeast infections. Focusing in more on the azoles specifically, we Azole Pharmacokinetics can see that several of these agents, such as Fluconazole Fluconazole Itraconazole Itraconazole Voriconazole Voriconazole Posaconazole Posaconazole Isavuconazole Isavuconazole posaconazole and itraconazole, require acid for Requires Requires acid acid for for No No Yes* Yes* No No Yes*** Yes*** No absorption No absorption absorption. This has been an Achilles heel for IV IV formulation formulation Yes Yes No No Yes** Yes** Yes Yes Yes Yes Protein Protein binding binding 10% 10% 99% 99% 58% 58% 99% 99% 99% 99% some of these azoles, and the newer Water Water solubility solubility Yes Yes No No No No No No Yes Yes formulations of posaconazole are sort of Renal Renal excretion excretion Yes Yes No No No No No No No No Active Active metabolites metabolites No No Yes Yes No No No No No No designed to address this issue. We also have the Kinetics Kinetics Linear Linear Linear Linear Nonlinear Nonlinear Linear Linear Linear Linear issue of non-linear pharmacokinetics that really Drug Drug interactions interactions *Capsules: absorption erratic, require acid and is best taken w/meals. Solution: best absorbed on empty stomach and less dependent stands out for voriconazole, and this can wreak *Capsules: absorption erratic, require acid and is best taken w/meals. Solution: best absorbed on empty stomach and less dependent on acidic ph on acidic ph **Contraindicated in patients with severe renal dysfunction due to toxicity of vehicle **Contraindicated in patients with severe renal dysfunction due to toxicity of vehicle ***Oral absorption enhanced by divided doses, high fat meals, avoidance of acid suppression therapy (less with H2RA) ***Oral absorption enhanced by divided doses, high fat meals, avoidance of acid suppression therapy (less with H2RA) all kinds of havoc for you when you re trying to Source: Manufacturer s prescribing information. Source: Manufacturer s prescribing information dose and do therapeutic drug monitoring for voriconazole, because just a simple, small dose change can result in dramatic changes in concentration with this agent. And we ll talk about that a little bit more after this section. In addition, you can see that for isavuconazole, it differs a little bit in that it is both available IV and orally, has good absorption, and has linear pharmacokinetics. So, there are a few properties there that make it stand out, and may have more reliable PK and therapeutic drug type of levels. Posaconazole Formulation & Target Thinking more about the posaconazole and these Concentrations new formulations, the delayed-release tablets Percent Percent Achieving Target Target really have changed how we can administer Median Median Trough Trough Level Level Concentrations 1, , posaconazole in the clinical setting. This oral 1, , , , suspension had a saturable absorption process, 70 1, , , ,000 which really made it difficult to get therapeutic drug concentrations in patients receiving this agent. The suspension is depicted here in green As you can see, when administered to patients, DR Tab Susp DR Tab Susp % > 700 ng/ml % > 1,250 ng/ml % > 700 ng/ml % > 1,250 ng/ml DR Tab Susp DR Tab Susp Durani U et al. Antimicrob Agents Chemother. 2015;59: only less than half could receive target Durani U et al. Antimicrob Agents Chemother. 2015;59: concentrations with this oral suspension. However, when switched to the delayed-release tablets, we can see that at least 70% of patients would achieve target concentrations for treatment of an invasive infection. The trough levels of the delayed release tablets, here in blue, are about twice that of the suspension formulation. And so, most centers have really switched over to using the delayed-release tablets, because you get much more reliable absorption. 25

27 Food Interactions with Azoles Impact Impact of of Food Food on onazole Concentrations Concentrations Administration Administration Recommendation Recommendation Fluconazole Fluconazole None None With With or or without without food food Itraconazole Itraconazole Capsules: Capsules: absorption absorption enhanced enhanced with with food food // acidic acidic beverage, beverage, reduced reduced by by agents agents that thataffect affect gastric gastric ph ph With With a a full fullmeal Solution: Solution: absorption absorption (~25%) (~25%) with with food food Empty Empty stomach stomach at at least least 11 hour hour before before food food Voriconazole Voriconazole Oral Oral absorption absorptiondelayed with with food food ( ( AUC AUC 22%) 22%) 11 h h before before or or after after food food Suspension: Suspension: absorption absorption by by high high fat fat meal, meal, liquid liquid With With full full meal, meal, or or liquid nutritional liquid nutritionalsupplement, and andacidic acidic carbonated carbonated nutritional nutritional supplement, supplement, or beverage or beverage ; ; impacted impacted by by agents agentsthat that affect affect gastric gastric ph Posaconazole ph Posaconazole acidic acidic carbonated carbonated beverage and beverage and GI GI motility motility DR DR tablets: tablets: absorption absorption enhanced enhanced by by high high fat fat meal meal With With food food Isavuconazole Isavuconazole Little Little impact impact of of high highfat fat meal meal (( AUC AUC 9%) 9%) With With or or without without food food In terms of food interactions, there are some with the azoles, and I ve just listed this summary slide here for you as a guide when dosing these in the clinical setting. Some do require to be given with food and others without food, and some require fat for absorption. So, this is important to remember when optimizing PK and doses for your patients. Chau MM et al. Intern Med J. 2014;44: Chau MM et al. Intern Med J. 2014;44: Cresemba Prescribing Information. Northbrook, IL: Astellas Pharma US; Cresemba Prescribing Information. Northbrook, IL: Astellas Pharma 51US; These drugs differ in terms of drug-drug Drug-Drug Interactions interactions as well. Polyenes, because they don t go through the P450 system for Polyenes metabolism, really have limited drug Limited to overlapping toxicity / compatibility interactions, primarily in the areas of Echinocandins overlapping toxicity, such as nephrotoxicity or Caspofungin: reduced by CYP450 inducers electrolyte imbalances, and compatibility issues Micafungin: weak inhibitor of CYP450 3A4 with other IV products. Echinocandins with caspofungin, you may see that it s reduced by Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S39. P450 inducers such as rifampin. This is really a Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S minor issue and can be managed often in the clinical setting. And then micafungin is a weak inhibitor of P450 3A4, and this can lead to some interactions with sirolimus and nifedipine, and those agents need to be monitored more closely when given with micafungin. Anidulafungin doesn t really go through the P450 system, and isn t limited by a chemical degradation problem, so anidulafungin does not have these interactions. 26

28 Cytochrome P450 & PgP Effects with Azoles Site Site Fluconazole Itraconazole Voriconazole Posaconazole Isavuconazole CYP3A4 CYP3A4 Inhibitor Inhibitor Substrate Substrate CYP2C8/9 CYP2C8/9 Inhibitor Inhibitor Substrate Substrate CYP2C19 CYP2C19 Inhibitor Inhibitor Substrate Substrate PgP PgP Inhibitor Inhibitor Substrate Substrate Chau MM et al. Intern Med J. 2014;44: Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S39. Cresemba Chau MM et al. Intern Med J. 2014;44: Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S39. Cresemba Prescribing Information. Northbrook, IL: Astellas Pharma US; Prescribing Information. Northbrook, IL: Astellas Pharma US; Interactions really come into play with the azole class. And as you can see here, I ve delineated which P450 isoenzyme sites are the major problems for these different agents. Voriconazole is notorious for 3A4 as well as 2C9 and 2C19 interactions, because it is both a substrate and inhibitor at this isoenzyme site. Most of the azoles do have P450 3A4 interactions, whereas 2C19 and 2C9 are predominantly with voriconazole. We re also getting more information on these agents and how they interact with transporters such as P-glycoprotein (PgP). As you can see here, posaconazole as well as itraconazole have PgP interactions, and they are both substrates and inhibitors of PgP. We re getting more information on this, and I think that we ll see more data in the future about these types of transporter interactions. But isavuconazole does not appear to have significant interactions other than the P450 3A4 issues. Looking at that a little bit more closely, if we re Effects of Newer Azoles on Immunosuppressants comparing the newer azoles, voriconazole still Voriconazole has probably the most significant interactions Posaconazole Isavuconazole CYP3A4 substrate (tablet) (oral (oral suspension) (capsule) with other immunosuppressants. Looking at Midazolam (3A4 (3A4 Probe) sirolimus, it will increase it by 11-fold, and Cyclosporine posaconazole also increases sirolimus levels to a Sirolimus similar degree. Isavuconazole has much less of Tacrolimus an interaction here, increasing sirolimus by twofold, and tacrolimus or cyclosporine one to Townsend R et al. Drug interaction profiles of isavuconazole, voriconazole and posaconazole with Townsend R et al. Drug interaction profiles of isavuconazole, voriconazole and posaconazole with immunosuppressants metabolized by CYP4503A4 (CYP3A4). Poster (P0216) presented at: 25th ECCMID; immunosuppressants metabolized by CYP4503A4 (CYP3A4). Poster (P0216) presented at: 25th ECCMID; two times. So, these interactions are thought to April 25-28, 2015; Copenhagen, Denmark. April 25-28, 2015; Copenhagen, Denmark be more manageable, and may be a differentiating feature for patients that require these agents to be given concomitantly. 27

29 In addition, the side effects really separate Common Side Effects of Antifungals these drugs. Polyenes are notorious for their nephrotoxicity, electrolyte imbalances, and Polyenes Echinocandins Azoles Hepatic other renal or infusion-related reactions that Renal Renal to to you can have. With echinocandins, you can see Hematologic + + NR NR Infusion-Related some infusion-related reactions when they are to to Electrolyte to to administered rapidly, but echinocandins are CNS CNS (vori) (vori) generally very well-tolerated. Azoles, on the Other Other differentiating features: Long Long term term toxicity, skin skin cancer, HCC HCC other hand generally, are generally welltolerated. You may see some LFT elevations, QTc QTcprolongation Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S39. Dodds Ashley ES et al. Clin Infect Dis. 2006;43:S28-S particularly in patients with hepatic comorbidities like hepatitis C, for example, or other hepatotoxic agents on board. And voriconazole really stands out, because we see side effects such as skin effects, ocular effects, and central nervous system effects with this agent that are different than the other azoles and can sometimes make it challenging to manage. Other differentiating features are some concerns about long-term toxicity with these agents. For those of you that are in transplant centers, or centers treating oncology patients, these patients might end up receiving an agent like voriconazole for two or three years, and there have been emerging concerns about genotoxicity of these agents that can lead to skin cancers as well as hepatocellular carcinomas. And so, there are some concerns about that, and we don t know probably enough about some of the newer agents in regards to that long-term toxicity, but we will be monitoring that in the future. QTc prolongation is another issue that often comes up. Voriconazole has been associated with that, as well as posaconazole. You can also see it with itraconazole. And interestingly, isavuconazole in the clinical trials was not shown to increase QTc interval in patients, and that may be another differentiating feature of that newer agent and might give particular advantage for some patients. Looking closer, specifically at invasive candidiasis Comparative Clinical Trials: FDA Indications and how we treat these infections, and looking AmB/ at the available agents, you can see that we have AmB/ Infection LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa a lot of clinical trials data for several different Invasive Candidiasis antifungal agents. It may surprise you that Aspergillosis salvage salvage salvage salvage Mucormycosis fluconazole does not actually have an FDA * * Prophylaxis indication for treatment of candidemia or Neutropenic Fever Fever invasive candidiasis. That was shocking to me * when I started preparing for this talk, because * No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): we use it so frequently. I think that it s just that Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): the company never applied for it. They do have indications for esophageal candidiasis and other oropharyngeal diseases. And it is very effective in clinical trials if you have a susceptible strain. We ll talk about that a little bit more in the future how it s positioned. 28

30 IDSA Guidelines: Candidiasis First First Line Line Alternative Alternative or or stepdown stepdown Alternative Alternative Echinocandin Fluconazole LFAB Candidemia Pappas PG et al. Clin Infect Dis. 2016:62(4):e1-e50. Pappas PG et al. Clin Infect Dis. 2016:62(4):e1-e Minimum Minimum duration: duration: 22 weeks weeks after after documented clearance clearance from from bloodstream The IDSA guidelines, based on recent trials data comparing these different agents that all have data, have put echinocandins as the first-line recommended agents for treatment of invasive candidiasis and candidemia, primarily because there s one study showing potentially a survival advantage; others showing general good efficacy, as well as excellent tolerability. And echinocandins do have activity against most of these potentially resistant strains that are sort of a problem with some of the azoles. Alternatively, they state, if a patient is clinically stable or is not likely to have an azole-resistant strain, then they may receive fluconazole as an alternative. Fluconazole is also possible as a stepdown therapy, once you know that the patient has a susceptible isolate. Another alternative would be lipid formulations of amphotericin B, and these agents are certainly going to be effective, and they may be even good choices if you believe a patient has an azoleresistant strain of infection. It s just the tolerability issues that lead them to be further down the list. For neutropenic patients, the guidelines switch it up just slightly and put lipid formulations of amphotericin B above fluconazole on this list. And the reason for that is that most of these patients with neutropenia have received azoles in the past or they were receiving azole prophylaxis at the time they had a breakthrough candidemia, and therefore they don t want to use fluconazole first-line until they re sure that a patient has a susceptible strain. Very importantly, the minimum duration of therapy for anybody with candidal bloodstream infections is at least two weeks after they clear the bloodstream. I cannot emphasize this point enough, that patients must receive follow-up blood cultures before you discharge them, ensuring that they ve cleared this infection; that they don t have some other ongoing source of infection. So, I really point that out, because that s an area where I think pharmacists can really help and ensure that patients have documented clearance before they head out the door. 29

31 So, the IDSA has come up with a checklist of Management of Candidemia sorts that I ve given you here, and other people Susceptibility have examined this sort of list as a bundle that testing you can look at to improve process measures for Follow-up cultures outcomes in patients with candidemia. So, on Dilated ophthalmologic exam this list are the follow-up cultures that I CVC removal mentioned; as well as the IDSA now Evaluate other sites recommends susceptibility testing for all patients Stepdown/duration of of therapy with fungal infections of the yeast variety with Candida. This can be challenging, I think, in the Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Antworth A et al. Pharmacotherapy. 2013;33(2): Antworth A et al. Pharmacotherapy. 2013;33(2): community hospital setting. I spend a lot of time in community hospitals, and there is a delay often. You have to do a send-out; it comes back a week or two later; and by that time the patient s already getting better and out the door. I think it s important to get a sense of the species that are common in your institution, and perhaps on a batch basis look at this, or carefully monitor this in your setting to see if that s appropriate. But due to the rates of resistance that we ve seen, susceptibility testing can be important. Other things that are important for these patients are a dilated ophthalmologic exam, removal of central venous catheters whenever possible, as well as evaluation, as I mentioned, of other sites of infection. Hepatosplenic candidiasis can linger and lurk in patients, and you might need to do imaging to evaluate that once neutrophils have recovered. And consider stepdown and overall duration of therapy based on the progress of the patient, and whether or not they clear this infection. But certainly, stewardship programs can have a huge role here in ensuring that patients meet these criteria and these procedures are performed appropriately before they leave the hospital. Predictors of Success This is some of the data that supports echinocandins as first-line options for IDSA. This was a meta-analysis performed by David Andes, 7 randomized controlled clinical trials, 978 and he looked at seven randomized controlled subjects clinical trials for candidemia in almost 1,000 Factor APACHE II II CVC removed Echinocandin OR (95% CI) 0.94 ( ) 1.69 ( ) 2.33 ( ) P value subjects and found that echinocandins were associated with a twofold increase in success. This was also correlated with a decrease in mortality and was significant for both albicans and non-albicans species. So, there seems to be Andes DR et al. Clin Infect Dis. 2012;54(8): Andes DR et al. Clin Infect Dis. 2012;54(8): something maybe with the cidal nature of echinocandins that allow them to be better than azoles as upfront therapy for these candidemic patients. 30

32 Comparative Clinical Trials: FDA Indications Infection AmB/ AmB/ LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa Invasive Candidiasis Aspergillosis salvage salvage salvage salvage Looking at aspergillosis, we have much less data here. As I said, for caspofungin we really only have salvage data; but we have clinical trials data for voriconazole, isavuconazole and lipid formulations of amphotericin B. Mucormycosis * * Prophylaxis Neutropenic Fever Fever * * No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Looking at the IDSA guidelines, they really IDSA Guidelines: Aspergillosis recommend voriconazole as first-line therapy. Invasive Pulmonary Aspergillosis This is based on the Herbrecht trial, which shows a documented mortality benefit of First First Line Line Voriconazole Treatment voriconazole over lipid formulations of duration: duration: weeks weeks L-AmB or or LFAB May May consider consider amphotericin B for patients with invasive Alternative secondary Alternative secondary prophylaxis Isavuconazole pulmonary aspergillosis. So, the alternatives would be the lipid formulations of amphotericin Alternative Alternative Combination: Voriconazole + Echinocandin B, which had about a 13% worse outcome than voriconazole in those trials; or isavuconazole, Patterson TF et al. Clin Infect Dis. 2016;63(4):e1-e60. Patterson TF et al. Clin Infect Dis. 2016;63(4):e1-e which we re going to talk about in a moment, which had similar efficacy as voriconazole in a comparative study; or a combination of voriconazole plus an echinocandin. Overall treatment duration should be at least six to twelve weeks, and these patients may require secondary prophylaxis if they have ongoing immunosuppression. 31

33 Looking at the data behind this, with the Combination Therapy voriconazole versus voriconazole plus Outcomes anidulafungin therapy, you can see that clinical p=0.077 outcomes were similar whether patients were 40 p= p=0.087 p= given monotherapy or combination therapy at six weeks and twelve weeks. There was certainly a trend towards success in the weeks 12 weeks Success 6 weeks 12 weeks Success combination therapy arm in terms of Vori Vori + Anid Vori Vori + Anid All-cause All-cause Mortality DRC DRC assessed Mortality assessed mortality a lower mortality. However, this did ITT mitt ITT mitt not reach statistical significance. Clinical success Marr KA et al. Ann Intern Med. 2015;162: Marr KA et al. Ann Intern Med. 2015;162: was based on a combination of complete and partial response. There were more patients in the combination therapy arm that were stable, and therefore didn t meet as much clinical success as the voriconazole that had more patients with a complete or partial response. But essentially, this was a negative trial. It did not find any differences that were statistically significant, and that s why you ll see combination therapy be positioned lowest in the guideline, behind voriconazole and those other formulations. Isavuconazole vs. Voriconazole Outcomes Day 42 Day 84 Success Day 42 Day 84 Success ISA Vori ISA Vori All-cause All-cause Mortality Mortality DRC DRC assessed assessed ITT ITT mitt mitt Looking at isavuconazole versus voriconazole in the SECURE study for invasive pulmonary aspergillosis, it had similar clinical outcomes in terms of mortality as well as success for both isavuconazole, in blue, and voriconazole, in green; and there were no statistically significant differences in terms of efficacy for these drugs in this trial. Maertens JA et al. The Lancet. 2016;387(10020): Maertens JA et al. The Lancet. 2016;387(10020): Isavuconazole vs. Voriconazole Adverse Adverse Events Events Skin Eye Hepatobiliary Skin Eye Hepatobiliary Isa Vori Isa Vori Maertens JA et al. The Lancet. 2016;387(10020): Maertens JA et al. The Lancet. 2016;387(10020): However, there were some differences in terms of adverse events (AEs). Overall, the patients had a similar rate of adverse events. However, if you hone in on three key adverse events, which were skin effects, ocular effects, and hepatobiliary adverse events, there was a statistically significant difference for all three of these, favoring isavuconazole. Isavuconazole had lower rates of AEs for each of these three, and in total that sums up about a 20% lower rate of adverse events with isavuconazole in this trial than voriconazole. 32

34 Comparative Clinical Trials: FDA Indications Infection AmB/ AmB/ LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa Invasive Candidiasis Aspergillosis salvage salvage salvage salvage Mucormycosis * * Prophylaxis Neutropenic Fever Fever * * No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Moving on to mucormycosis, we have even less data here. We have no randomized controlled trials. These are difficult-to-treat patients, and we are unlikely to have a big trial of this any time soon. So, what we re left with are some sort of series case series of patients that have received isavuconazole or lipid formulations of amphotericin B or conventional amphotericin B. Rare Mold Infections: Mucormycosis Identification to species level Susceptibility testing Imaging Histopathology What s important here is what we mentioned earlier: identify these to a species level, because sometimes labs can get it wrong. Make sure that you do susceptibility testing, because the drugs can differ in terms of activity. Proceed with imaging and histopathology to really get a good diagnosis. Cornely OA et al. Clin Microbiol Infect. 2014;20(Suppl 3):5-26. Cornely OA et al. Clin Microbiol Infect. 2014;20(Suppl 3): There are no United States guidelines from IDSA Rare Mold Infections: Mucormycosis to help us manage these. However, the First-Line Other options Europeans do have guidelines, and I saw a few Surgery Isavuconazole Europeans in the room getting breakfast this Liposomal Combination therapy Amphotericin B ( ( 5 morning, so I m glad you re here with us. mg/kg/day) Note: Voriconazole lacks activity Salvage What s recommended really is surgery. Posaconazole Generally, these patients do better if you can cut this fungal infection out of the tissue, and then use adjunctive antifungal therapy, perhaps with liposomal amphotericin B; salvage might be an Cornely OA et al. Clin Microbiol Infect. 2014;20(Suppl 3):5-26. Cornely OA et al. Clin Microbiol Infect. 2014;20(Suppl 3): option for posaconazole or combination therapy upfront; and isavuconazole does have an FDA indication for this and was released after these guidelines were published. So, we ll be looking forward to new versions of these guidelines and seeing how they incorporate that new option. Of note, it s important just to remember that voriconazole lacks activity here. 33

35 Comparative Clinical Trials: FDA Indications Now we re just going to focus a little bit on prophylaxis and neutropenic fever. Infection AmB/ AmB/ LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa Invasive Candidiasis Aspergillosis salvage salvage salvage salvage Mucormycosis * * Prophylaxis Neutropenic Fever Fever * * No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Prophylactic & Empiric Therapy Amphotericin B Posaconazole Voriconazole Isavuconazole Micafungin Caspofungin Fluconazole Anidulafungin Yeast Yeast Mold Mold And for both of these, what you re going to use in terms of fungal therapy depends on the risk of your patient. Are they more at risk of yeast? Are they more at risk of molds? And I ve organized the agents here in terms of that spectrum, knowing that echinocandins lack the cryptococcal activity that fluconazole has, that covers both Candida and Crypto, so it has the broadest yeast activity, and the agents with more broad mold activity are here to the right Prophylaxis Pathogen Pathogen Setting Setting Recommendations Candida Candida High High risk risk patients patients in in ICUs ICUs with with high high rate rate (>5%) (>5%) Fluconazole Fluconazole Alternative: Alternative: echinocandin Chlorhexidine baths baths Candida/ Candida/ molds molds Candida Candida Candida/ Candida/ molds molds HSCT HSCT (GVHD)/hematologic malignancy malignancy High High Risk Risk Hematologic Hematologic malignancy malignancy Lower Lower Risk Risk Lung Lung transplant transplant Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Fleming S et al. Inter Med J. 2014;44(12b): Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Fleming S et al. Inter Med J. 2014;44(12b): Flowers CR et al. J Clin Oncol. 2013;31(6): Freifeld AG et al. Clin Infect Dis. 2011;52(4):e56-e93. Flowers CR et al. J Clin Oncol. 2013;31(6): Freifeld AG et al. Clin Infect Dis. 2011;52(4):e56-e93. Tacke E et al. Ann Hematol. 2014;93(9): Husain S et al. J Heart Lung Transp. 2016;35(3): Tacke E et al. Ann Hematol. 2014;93(9): Husain S et al. J Heart 70 Lung Transp. 2016;35(3): Posaconazole Other Other azoles, azoles, candins, candins, AmB/LFAB AmB/LFAB also also acceptable acceptable Fluconazole Fluconazole Other Other azoles, azoles, candins candinsalso also acceptable acceptable Azole Azole + + inhaled inhaled LFAB LFAB We do have emerging data in the prophylaxis area. And again, depending on which organism you re targeting, would guide your approach. So, Candida prophylaxis is now only recommended in patients who are in the ICUs, where there is at least a 5% incidence of candidemia or invasive candidiasis. We do have those diagnostic issues, but if you do believe that your ICU has a high rate, then fluconazole prophylaxis may be of benefit there. In stem cell transplant patients and those with hematologic malignancies, the prophylaxis strategy depends on the risk. Again, if they re higher-risk, with myeloablative agents, or lowerrisk, where you can use posaconazole first-line or perhaps fluconazole in lower-risk patients. And then for lung transplant patients, not to forget the solid organs, azoles plus inhaled lipid formulations of amphotericin B are commonly used in a lot of centers. 34

36 Comparative Clinical Trials: FDA Indications Infection AmB/ AmB/ LFAB LFAB Anid Anid Cas Cas Mica Mica Flu Flu Itra Itra Vori Vori Posa Posa Isa Isa Invasive Candidiasis Aspergillosis salvage salvage salvage salvage Finally, for neutropenic fever, this is an area of a lot of study. We ve had several randomized controlled trials with these agents, as well as several with FDA indications. Mucormycosis * * Prophylaxis Neutropenic Fever Fever * * No randomized, comparative clinical trials No randomized, comparative clinical trials Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Wingard JR et al. Blood Dec 9;116(24): Winston DJ et al. Am J Med. 2000;108(4): Boogaerts M et al. Ann Intern Med Sep 18;135(6): Boogaerts M et al. Ann Intern Med Sep 18;135(6): This is a busy slide, but I just want to point out a Empiric Therapy Neutropenic Fever: IFI Rates few key points. Without antifungal empirical Ref Ref Year Year Abx Abx Fluc Fluc Caspo Caspo Itra Itra Vori Vori AmB/LFAB AmB/LFAB therapy in the old trials, we saw an incidence of Pizzo Pizzo % 31% 6% 6% EORTC EORTC % 9% 1% 1% invasive fungal infections up to 30% in patients AmB 1 L-AmB 1 L-AmB 3 AmB 1 L-AmB 1 L-AmB 3 Prentice Prentice 2% 2% 2.6% 2.6% 1.6% 1.6% with neutropenic fever. We were able to AmB ABCD AmB ABCD White White % 14.3% 14.7% 14.7% AmB L-AmB decrease that anywhere between 1 to 6% AmB L-AmB Walsh Walsh % 8% 3% 3% Winston incidence with antifungal therapy. So, this is a Winston % 8% 6% 6% ABLC 5 L-AmB 3 L-AmB 5 ABLC 5 L-AmB 3 L-AmB 5 Wingard Wingard % key area where antifungals can be administered 3.8% 3.6% 3.6% 2.5% 2.5% Boogaerts Boogaerts % 3% 3% 3% Walsh to neutropenic patients. Choosing between the Walsh % 1.9% 5% 5% Walsh Walsh % 5% 4% 4% agents is a little bit difficult because these Adapted from Busca A. Med J Hematol Infect Dis Vol 8. Adapted from Busca A. Med J Hematol Infect Dis Vol randomized trials all show them to be pretty equivalent. What was not equivalent was voriconazole, which was not non-inferior in some of these clinical trials. So, lipid formulations of amphotericin B may be acceptable here, as well as echinocandins; and some shops may also use an azole. The IDSA does have some guidelines here, and Empiric Therapy for neutropenic fever they list all of these Neutropenic Fever No No prophylaxis Caspofungin choices, and UpToDate has a nice guide that AmB AmB LFAB LFAB Fluconazole helps you, depending on whether a patient has Caspofungin Mold Mold active active agent prophylaxis agent Itraconazole Voriconazole been receiving prophylaxis or not, what to Mold Mold Alternative class class prophylaxis mold mold active active agent agent choose. Because obviously if a person has been ICU/Non-neutropenic Recommended Echinocandin receiving fluconazole prophylaxis, you don t AmB AmB No No recent recent azole, azole, not LFAB not LFAB colonized want to start out with that as your agent of colonized w/ w/ Fluconazole Echinocandin resistant resistant Candida Candida Fluconazole choice when they get a fever. So, this helps Alternative LFAB LFAB guide you. If they ve been on azole prophylaxis Freifeld AG et al. Clin Infect Dis. 2011;52(4):e56-e93. Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Wingard JR. Treatment of Freifeld AG et al. Clin Infect Dis. 2011;52(4):e56-e93. Pappas PG et al. Clin Infect Dis. 2016;62(4):e1-e50. Wingard JR. Treatment of neutropenic fever syndromes in adults with hematologic malignancies and hematopoietic cell transplant recipients. UpToDate neutropenic fever syndromes in adults with hematologic malignancies and hematopoietic 73 cell transplant recipients. UpToDate with fluconazole, you would then use a moldactive agent to treat them when they break through with fever. Likewise, if they ve been on mold prophylaxis, then you would use an alternative class of a mold-active agent. 35

37 For ICU patients or non-neutropenic patients, we have these choices. And again, your selection may be based on whether or not they had received a recent azole, in which case, if they have not, then you could proceed with fluconazole; and if they had, you might use an alternative agent. Summary of Treatment/Prophylaxis Risk and treatment strategies vary depending on clinical factors Organism- and infection site-specific approach Evidence-based guidelines can help Stewardship/monitoring to optimize outcomes Finally, we know that the risk and treatment strategies will vary depending on clinical factors, and we need to take organism and site-specific approaches. These evidence-based guidelines can help us, and we can use our stewardship and monitoring through our pharmacy friends here to optimize patient outcomes. Thank you. And I m going to turn it back over to Ryan

38 I want to start by asking another question of you, and this is about therapeutic drug monitoring. What is is the goal voriconazole trough concentration in in a patient being treated for for invasive aspergillosis? Audience Response What is the goal voriconazole trough concentration in a patient being treated for invasive aspergillosis? A wide variety of answers here. And I think you ll see in some of the subsequent slides, we ve now become pretty comfortable in recommending a trough concentration of at least 1 for voriconazole for the treatment of invasive fungal infections, which we ll talk about Personalizing Antifungal Treatment Review the expanding roles for antifungal susceptibility testing in in patient management Fluconazole and echinocandin testing against Candida Azole testing for Aspergillus Discuss challenges and opportunities in in therapeutic drug monitoring (TDM) of antifungals Who, what, when, and how of of antifungal TMD We talked a lot today about the epidemiology of invasive fungal infections; the diagnostic tests we can use; and then Dr. Johnson did a nice summary of the antifungal agents. I want to talk to you finally today about two really important avenues for pharmacists to be involved in the management of these invasive fungal infections, and that s, first, the role of antifungal susceptibility testing in the clinic; and then second, therapeutic drug monitoring of these drugs. 37

39 Antifungal Drug Susceptibility Testing (AFST) Antifungal drug resistance is is increasing Changing epidemiology of of Candida species Expanded use of of empiric and pre-emptive antifungals Growing at-risk populations In In vitro AFST is is becoming an important part of patient management and surveillance New agents are being introduced into the clinic Detection of of resistance helps prioritize use of of antifungals Antifungal susceptibility testing really has become very important because of the changing epidemiology in Candida species. Dr. Slain showed you that we re seeing increasing rates of resistance to Candida species specifically, and we know we have growing at-risk populations. So, because of this, antifungal susceptibility testing is really important in getting patients the right drug as soon as we can IDSA Recommendations Invasive candidiasis Testing for for azole susceptibility for for all all bloodstream and other clinically relevant Candida isolates Testing for for echinocandin susceptibility should be be considered in in patients who have had prior echinocandin treatment Also for for those who have infection with C. C. glabrata or or C. C. parapsilosis Aspergillosis Routine susceptibility testing of of isolates recovered during initial infection is is not recommended Pappas PG, et al. Clin Infect Dis. 2015;62:e1-50; Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60. Pappas PG, et al. Clin Infect Dis. 2015;62:e1-50; Patterson TF, et al. Clin Infect Dis. 2016;63:e So, the guideline recommendations are to test azole susceptibility for all bloodstream Candida isolates, and echinocandin susceptibility should be considered, particularly among patients with prior echinocandin exposure, and we ll talk about some of the roles for echinocandin susceptibility testing momentarily. The aspergillosis guidelines, however, do not recommend routine susceptibility testing, and we ll talk about reasons why. So, the nuts and bolts of susceptibility testing Nuts and Bolts of AFST are that it s important to understand that the Standardized methods developed by by the Clinical and CLSI, or Clinical and Laboratory Standards Laboratory Standards Institute (CLSI) for both Aspergillus and Candida species Institute, has really gone through a very Provide antifungal drug minimum inhibitory concentration laborious process to standardize the testing Broth microdilution and disc diffusion methods methodology for both Candida and MICs are interpreted by by clinical breakpoints that have aspergillosis. And what these standardized been developed based on antifungal PK-PD, MIC distributions, resistance mechanisms, and outcome data tests provide us is some measure of drug Goal is is to to predict the likely outcome of of therapy susceptibility, which we typically measure in drug minimum inhibitory concentration, or MIC; and then we can interpret these MICs based on clinical breakpoints, which have been developed based on extensive antifungal PK-PD data, MIC distributions, resistance mechanisms, and some outcome data. And the goal of these susceptibility testing methods is to be able to predict the outcome of your patient. 38

40 Candida AFST Clinical Breakpoints Starting with fluconazole, we now have some pretty reliable susceptibility breakpoints for Fluconazole Echinocandins Candida albicans, tropicalis and parapsilosis, but Species Species SS S-DD S-DD R Species Species SS II R bear in mind Candida krusei is intrinsically C. C. glabrata glabrata C. C. glabrata glabrata ** ** ** C. C. albicans albicans C. C. parapsilosis parapsilosis C. C. albicans albicans C. C. parapsilosis parapsilosis C. C. krusei* krusei* C. C. tropicalis tropicalis C. C. krusei krusei C. C. tropicalis tropicalis resistant, so you will not see susceptibility breakpoints for fluconazole there. Candida glabrata stands out as the exception here, Drug* Drug* SS II R because there is not a susceptibility breakpoint Anidula/Casp Micafungin Micafungin for Candida glabrata, only a susceptible dosedependent ** C. C. krusei kruseiconsidered to to be be intrinsically intrinsically resistant resistant Clinical Laboratory Standards Institute, Wayne PA. Document M27-S4 (Tables Clinical Laboratory Standards Institute, Wayne PA. Document M27-S4 (Tables interpretation, which will be 1 and 2) 1 and 2) 81 important as we move forward. Similarly, for the 81 echinocandins, we have a number of breakpoints that have been refined since their first adoption, and particularly refined with Candida glabrata, which we know is the problematic pathogen for echinocandin resistance, and we ll talk about that momentarily. Rates of Fluconazole Resistance ~7% ~7% of of Candida bloodstream isolates are are resistant to to fluconazole 14% 14% 12% 12% 11.9% 11.9% 10% 10% 8% 8% 6% 6% 6.2% 6.2% 4.1% 4.1% 4% 4% 2.3% 2.3% 2% 2% 0% 0% C. C. albicans albicans (n=877) (n=877) C. C. glabrata glabrata (n=670) (n=670) C. C. parapsilosis (n=389) (n=389) C. C. tropicalis tropicalis (n=241) (n=241) ** C. C. krusei kruseiisolates isolates considered intrinsically resistant resistant Lockhart SR, et al. J Clin Microbiol. 2012;50:3435. Lockhart SR, et al. J Clin Microbiol. 2012;50: Let s start with fluconazole. Again, as Dr. Slain pointed out, the big problem with both antifungal resistance is Candida glabrata, and this is where we see the highest rates of fluconazole resistance. So, I want to talk to you about, what s the role for fluconazole susceptibility testing? How can you use this test in your hospitals to help your patients? What Are the Roles for Fluconazole AFST? Some assumptions can be be made based upon identification of of the Candida species: C. C. albicans Usually susceptible C. C. krusei Intrinsically resistant Role #1: MICs are associated with patient outcomes C. C. albicans candidemia treated with fluconazole (n=217), rates of of Well, I think the first thing to understand is that you can make some assumptions based on what we know from large epidemiology databases, which is that Candida albicans is usually susceptible to fluconazole and Candida krusei is almost always intrinsically resistant. infection related mortality were: Fluconazole % N MIC MIC Success How do you use these things? Well, I think role 20.6% if if fluconazole Success MIC MIC >2 >2 µg/ml 92 number one is, we know that MICs for % if if fluconazole MIC MIC 2 µg/ml (P=0.001) 37 fluconazole are associated with patient outcomes. And in fact, this has been shown in van Hal SJ, et al. J Antimicrob Chemother. 2014;69: ; van Hal SJ, et al. J Antimicrob Chemother. 2014;69: ; Pfaller MA, et al. J Clin Microbiol. 2012;50: Pfaller MA, et al. J Clin Microbiol. 2012;50: very nice outcome data that s been put together comprehensively by Dr. Mike Pfaller. But you can look at the table here. As your fluconazole MIC increases, representing lower drug susceptibility, your percent of clinical success goes down. And these are fairly large numbers. So, we know having a low fluconazole MIC and treating that patient with fluconazole is associated with a better outcome than having a higher MIC. So, we know MIC is important for picking the right drug. 39

41 What Are the Roles for Fluconazole AFST? MIC is also important for dosing fluconazole. Role #2: #2: MIC-based fluconazole dosing matters! There is a wealth of literature out there which I m Endpoint Endpoint Acceptable Acceptabletarget target Optimal Optimal target target Efficacy Efficacy in in mouse mouse model model AUC/MIC AUC/MIC AUC/MIC AUC/MIC going to summarize here for you briefly. This has Clinical Clinicalsuccess in in patients patients Estimated Estimated AUC/MIC AUC/MIC Estimated Estimated AUC/MIC AUC/MIC 75 with with candidemia 75 candidemia been shown in both mouse models, clinical Probability Probability of of PK-PD PK-PD target target Daily Daily dose/mic dose/mic to to Daily Daily dose/mic dose/mic to to attainment attainment in in humans humans achieve achieve AUC/MIC AUC/MIC achieve achieve AUC/MIC AUC/MIC success in patients, and then some modeling with Fluconazole Fluconazole MIC MIC Daily Dailydose needed needed to to Candida PK-PD target attainment that have consistently Candidaspecies (Interpretation) achieve achieve AUC/MIC AUC/MIC (S) (S) 3mg/kg/day 3mg/kg/day (200mg) (200mg) C. C. albicans albicans 44 (S-DD) (S-DD) 6mg/kg/day 6mg/kg/day (400mg) (400mg) shown us for fluconazole the best target is an AUC 88 (R) (R) 12mg/kg/day 12mg/kg/day (800mg) (800mg) (S-DD) (S-DD) 24mg/kg/day 24mg/kg/day (1600mg) (1600mg) to MIC ratio of at least 50. You can see some of C. C. glabrata glabrata (S-DD) (S-DD) Not Not achievable achievable (R) (R) Not Not achievable achievable the estimates that have range here, and all the Pfaller MA, et al. J Clin Microbiol. 2012;50:2846; Rodriguez-Tudela JL, et al. Antimicrob Agents Chemother. 2007;51:3559; Pfaller MA, et al. J Clin Microbiol. 2012;50:2846; Rodriguez-Tudela JL, et al. Antimicrob Agents Chemother. 2007;51:3559; Clancy CJ, et al. Antimicrob Agents Chemother. 2005;49:3171; Pai M, et al. Antimicrob Agents Chemother. 2007;51:35; Clancy CJ, et al. Antimicrob Agents Chemother. 2005;49:3171; Pai M, et al. Antimicrob Agents Chemother. 2007;51:35; Baddley JW, et al. Antimicrob Agents Chemother. 2008;52:3022; Patel K, et al. Antimicrob Agents Chemother. Baddley JW, et al. Antimicrob Agents Chemother. 2008;52:3022; Patel K, et al. Antimicrob Agents Chemother. references are at the bottom of the slide if you 2011;55:5868; Han S, et al. Antimicrob Agents Chemother. 2013;57: ;55:5868; Han S, et al. Antimicrob Agents Chemother. 2013;57: want to do your own investigation. But the reason why that s important is because, when we start to look at the MICs of fluconazole, we know that we can clinically achieve some of these AUC to MIC targets if the MIC is low. For instance, with Candida albicans, if you have a MIC less than 2, you know at least 200 milligrams a day will typically give you an AUC to MIC ratio greater than 50. But as these MICs start to increase, 4 to 8 to even 16, you quickly see that we re giving very high doses of fluconazole. And in fact, as you get into the higher end of the susceptible dose-dependent range for Candida glabrata, these really aren t clinically-achievable fluconazole doses. So, fluconazole susceptibility testing is important because that MIC helps you decide what dose of fluconazole you need to give your patients; again, trying to target an AUC to MIC ratio of at least 50. Role number three for fluconazole this is What Are the Roles for Fluconazole AFST? particularly useful for patients who have had Role #3: Useful for patients with prior fluconazole prior exposure to antifungals. We know that exposure and those with recurrent infections patients that develop antifungal resistance are Rates of of prior fluconazole exposure across all all Candida: typically patients that have been previously Fluconazole susceptible (n=107) = 8% 8% Fluconazole non-susceptible (n=26) = 35% (P (P = 0.002) exposed to these agents. So, fluconazole testing Suboptimal dosing also contributes to to resistance is absolutely necessary for patients that have Rates of of fluconazole resistance increase significantly among been previously exposed to these antifungal persistent or or recurrent candidemia agents, and these are some data that are showing you just that that the rate of Shah DN, et al. Antimicrob Agents Chemother. 2012;56: Shah DN, et al. Antimicrob Agents Chemother. 2012;56: fluconazole non-susceptibility is 35% among patients with prior exposure, and only 8% among patients without prior exposure. 40

42 Rates of Echinocandin Resistance ~3% ~3% of of Candida bloodstream isolates are are resistant to to echinocandins 9% 9% 8% 8% 7.8% 7.8% 7% 7% 6% 6% 5% 5% 4% 4% 3% 3% 2% 2% 1% 1% 0.6% 0.6% 0% 0% 0.0% 0.0% 0.0% 0.0% C. C. albicans albicans (n=169) (n=169) C. C. glabrata glabrata (n=167) (n=167) C. C. parapsilosis parapsilosis (n=71) (n=71) C. C. tropicalis tropicalis (n=38) (n=38) ** 11 of of 66 C. C. krusei kruseiwas was resistant resistant (17%) (17%) Shields RK, et al. Antimicrob Agents Chemother. 2015;59:7465. Shields RK, et al. Antimicrob Agents Chemother. 2015;59: Let s transition then to echinocandin resistance. Again, the major problem with echinocandin resistance is Candida glabrata. Rates of FKS Mutant Candida And we know that for Candida glabrata, the overall rate ranges from 4 to 18% for echinocandin resistance, is much lower for Percentage Percentage of of echinocandin candidiasis candidiasis during during that that test test Species Species Overall Overall Any Any prior prior exposure* exposure* Breakthrough echinocandin therapy* therapy* FKS FKS mutants mutants resistant resistant C. C. albicans albicans 11 3% 3% 5% 5% 50% 50% ~100% ~100% Candida albicans, and rarely reported for the other species. The rate or incidence of resistance C. C. glabrata glabrata 44 18% 18% % 32% % 75% % 81% typically goes up with prior drug exposure, C. C. krusei krusei Case Case reports reports C. C. tropicalis tropicalis Case Case reports reports particularly patients who are on the drug and develop disease. Now, echinocandin resistance is *Both *Both the the duration duration and and timing timing of of prior prior echinocandin exposure exposure are are likely likely mediated by FKS mutations. FKS is a gene that associated associated with with the the frequency frequency of of FKS FKS gene gene mutations mutations RK, et al. Curr Opin Infect Dis. 2015;6:514. encodes some of these beta-glucans in the Shields Shields RK, et al. Curr Opin Infect Dis. 2015;6: Candida cell wall, and we know that for Candida 87 glabrata that the vast majority of resistant strains have FKS mutations, and both the duration and timing of prior drug exposure is important. So, why are we talking about FKS mutations? What Are the Roles for Echinocandin AFST? Well, role number one for echinocandin Role #1: Detect echinocandin-resistant C. C. glabrata susceptibility testing is to be able to detect Elevated echinocandin MICs and the presence of of FKS gene mutations are associated with clinical failures these echinocandin-resistant Candida glabrata, Variable Variable Failure Failure Rate Rate P-value P-value Reference Reference because we know echinocandin resistance is Caspofungin MIC MIC >0.5 >0.5 µg/ml µg/ml 60% 60% Shields RK, et al. Shields RK, et al. Prior Prior echinocandin exposure exposure 54% 54% Antimicrob Agents Antimicrob Agents Chemother 2012;56:4862 associated with worse patient outcomes, as is Presence Presence of of FKS FKSmutation 86% 86% Chemother 2012;56: Caspofungin MIC MIC µg/ml µg/ml 73% 73% NA NA Alexander BD, et al. Clin true for FKS gene mutations. This has been Alexander BD, et al. Clin Presence Presence of of FKS FKS mutation mutation 62% 62% Infect Dis 2013;56: Infect Dis 2013;56:1724 shown in studies that we ve done in the Caspofungin MIC MIC µg/ml µg/ml 47% 47% Beyda ND, et al. Clin Infect Beyda ND, et al. Clin Infect Echinocandin exposure exposure 73% 73% <0.01 <0.01 Dis 2014;59:819 Dis 2014;59:819 Presence Presence of of FKS FKSmutation 60% 60% University of Pittsburgh, as well as Dr. Johnson at Duke University, and Nick Beyda at the Shields RK, et al. Curr Opin Infect Dis. 2015;6:514. Shields RK, et al. Curr Opin Infect Dis. 2015;6: University of Houston consistently shows us that if you have an echinocandin-resistant Candida glabrata, that patient is associated with a worse outcome. So, we have to be able to detect these in the microbiology lab. 41

43 What Are the Roles for Echinocandin AFST? Role number two for echinocandin susceptibility testing is, as Dr. Johnson just showed you, that Role Role #2. #2. De-escalation of of antifungal treatment the echinocandins are now recommended as Echinocandins recommended as as initial initial therapy for for the the treatment of of candidemia front-line therapy for invasive candidiasis. But it De-escalation to to fluconazole recommended among patients who who have: have: Improved Improved clinically clinically on on echinocandin therapy therapy doesn t necessarily mean that you have to treat Documented clearance clearance of of Candida Candidafrom blood blood Infected Infected with with an an organism organism susceptible susceptible to to fluconazole fluconazole that patient for 14 days with an echinocandin. Open-label study study comparing anidulafungin and and These agents are more expensive than azoles. 100% 100% 80% 80% anidulafungin, then then deescalation to to fluconazole de- They also require intravenous access. So, there is 60% 60% 40% 40% De-escalation deemed a safe safe and and effective approach potentially a role for de-escalation of antifungal 20% 20% 0% 0% Anidulafungin De-escalation therapy, and in order to de-escalate from an Anidulafungin De-escalation Global Clinical Microbiologic Global Clinical Microbiologic Pappas PG, et al. Clin Infect Dis. 2015;62:e1-50; Vazquez J, et al. BMD Infect Dis. 2014;14:97. echinocandin to an azole agent you need to be Pappas PG, et al. Clin Infect Dis. 2015;62:e1-50; Vazquez J, et al. BMD 89 Infect Dis. 2014;14: able to assure your clinicians that that isolate is susceptible to fluconazole. And there have been very nice data collected by Jose Vazquez, showing that de-escalation strategies are associated with just as good clinical outcomes as using an echinocandin for the duration of therapy. Issues with Echinocandin AFST #1. Unclear relevance outside of of C. C. glabrata Echinocandin resistance and FKS mutations are are rare among non-c. glabrata species #2. Echinocandin MICs vary between agents Caspofungin > anidulafungin > micafungin Significant inter-laboratory variability for for caspofungin #3. Clinical laboratories don t use reference methods Sensititre YeastOne panels overcall caspofungin resistance against C. C. krusei and C. C. glabrata Etest, Vitek-2, and disc diffusion methods also used But there is some hesitation with echinocandin susceptibility testing, which is important for you to be aware of. The reason why this isn t uniformly recommended and the first issue with echinocandin susceptibility testing is that there s unclear relevance outside of Candida glabrata. We really only see resistance for Candida glabrata, so why test all the other species? That s something that remains to be worked out. Shields RK, et al. Curr Opin Infect Dis. 2015;6:514; Shields RK, et al. Curr Opin Infect Dis. 2015;6:514; Eschenauer GA, et al. Antimicrob Agents Chemother. 2014; 58:1897. Eschenauer GA, et al. Antimicrob Agents Chemother. 2014; 58: Echinocandin MICs also vary between the agents. So, depending on the testing platform you re using at your hospital, it may or may not even include the formulary echinocandin that you use. Typically, MICs are higher for caspofungin and lower for micafungin. And particularly for caspofungin, there s significant inter-laboratory variability with testing this agent. Issue number three is that the clinical laboratories don t actually use the reference methods. The reference methods are laborious and labor-intensive, they re expensive, and take time. So, they typically use automated systems like the Sensititre YeastOne panels, maybe Etest or Vitek, which are not the CLSI-recommended methods, and it s important because some of these methods do overcall resistance for Candida glabrata and Candida krusei specifically. 42

44 A practical approach, and something that we try A Practical Approach for Echinocandin AFST to do at the University of Pittsburgh, is think Prior Prior echinocandin exposure exposure about how you can use these tests rationally in *Validated *Validated by by local local Yes the clinic. And for us, what we know is important Yes epidemiology and No and No testing testing methods methods for echinocandin resistance is the top of our Echinocandin MIC MIC Echinocandin MIC MIC classified classified as as resistant* resistant* classified classified as as resistant* resistant* diagram here. If your patient has prior Yes Yes No No Yes Yes No No echinocandin exposure, they re at risk for Probability resistance. If they don t, they are not really at Probability of of treatment treatment Probability Probability of of treatment treatment failure failure failure failure risk for resistance. So, we can use this as a kind >50% >50% 30-50% 30-50% 30-50% 30-50% <30% <30% of decision tree and say, if they have prior Shields RK, et al. Curr Opin Infect Dis. 2015;6:514. Shields RK, et al. Curr Opin Infect Dis. 2015;6: exposure, then we need to know what the MIC is, and we ll classify it as resistance, and this has to be classified as resistance based on the local testing method that you re using. So, what method are you using? What s your MIC distribution at your hospital? Because there are some issues with breakpoints in the echinocandins still. So, if the isolate is classified as resistant by your local methods and epidemiology, we know the probability of treatment failure is highest in that patient; and on the other end of the spectrum, if they have no prior exposure and the MICs are low, they have the lowest probability of failures. Here s an idea of how you can use some of these tests in your hospital. Aspergillus AFST Clinical breakpoints have not been developed by by CLSI Low rates of of azole resistance in in the United States, but may be be increasing worldwide Across US US centers: <1% <1% azole resistance among A. A. fumigatus No No correlation between MICs and and mortality Worldwide collection: 1.4% voriconazole resistance against A. A. fumigatus No clear indication that MICs influence outcomes AFST reserved for patients unresponsive to to ongoing treatment or or in in whom antifungal resistance is is suspected Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Baddley JW, et al. J Clin Microbiol. 2009;47:3271; Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Baddley JW, et al. J Clin Microbiol. 2009;47:3271; Pfaller MA, et al. J Clin Microbiol. 2011;49:586. Pfaller MA, et al. J Clin Microbiol. 2011;49: Briefly, on Aspergillus antifungal susceptibility testing (AFST), there are not clinical breakpoints by the CLSI. Dr. Slain pointed out that resistance is increasing in Europe. It s been less so in the US, but I think it s something that s coming and we need to be aware of. Until we see more resistance, there really won t be any clear indications that MICs influence outcomes, because we typically see mostly susceptible strains. 43

45 Personalizing Antifungal Treatment Review the expanding roles for antifungal susceptibility testing in patient management Fluconazole and echinocandin testing against Candida Azole testing for Aspergillus Discuss challenges and opportunities in therapeutic drug monitoring (TDM) of antifungals Who, what, when, and how of antifungal TMD 93 TDM: Who? Recommended for patients who receive: Azole-based therapy for Aspergillosis Antifungal prophylaxis with azoles Concomitant agents that may interact with antifungals Particularly important for patients with: Pharmacokinetic variability (ICU, obese, etc.) Severe fungal infections Suspected toxicity or or breakthrough infections Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60. Patterson TF, et al. Clin Infect Dis. 2016;63:e Briefly, as we finish up, I want to talk about the who, what, where, and why of antifungal therapeutic drug monitoring (TDM). This is particularly important for patients that receive azole-based therapy for aspergillosis as well as antifungal prophylaxis. And it s even more important for our special populations, and those are populations that have pharmacokinetic variability, such as the obese or ICU patient populations; those with fungal infections or suspected toxicity. TDM: What drugs? Antifungal Antifungal Drug Drug Poor Poor oral oral Inter-patient Inter-patient Exposure: Exposure: Exposure: Exposure: Drug-drug Drug-drug bioavailability bioavailability variability variability outcome outcome toxicity toxicity interactions interactions Amphotericin B No No Yes Yes Yes Yes No No Fluconazole Fluconazole No No No No Yes Yes Yes Yes Yes Yes Isavuconazole No No No No Unknown Unknown Unknown Unknown Yes Yes Itraconazole Itraconazole Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Voriconazole No No Yes Yes Yes Yes Yes Yes Yes Yes Posaconazole Suspension Suspension Tablet Tablet Intravenous Intravenous Yes Yes No No Yes Yes No No No No Yes Yes Yes Yes Yes Yes No No Unknown Unknown Unknown Unknown Yes Yes Yes Yes Yes Yes Echinocandins No No Yes Yes No No No No Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2:55. Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2: Typically, what we re looking for in drug properties that require us to monitor serum drug levels is a number of characteristics. Does the drug have poor bioavailability? Is there interpatient variability? Is exposure to that drug associated with some clinical outcome? Also, is exposure associated with toxicity of that drug? And then, are there drug-drug interactions? And really, when you consider these characteristics, there s three agents that we should be monitoring, and those are itraconazole, voriconazole, and the posaconazole suspension. 44

46 TDM: When and What Target? Antifungal Antifungal Drug Drug Time Time to to steady-state steady-state Goal Goal trough trough concentration for for prophylaxis prophylaxis Goal Goal trough trough concentration for for treatment treatment Itraconazole Itraconazole ~ ~ 77 days days Posaconazole days days Voriconazole* days days *RCT *RCT of of voriconazole voriconazole treatment treatment with with or or without without TDM TDM showed showed clinical clinical benefit benefit for for TDM TDM among among patients patients with with IFIs IFIs Trough concentrations should be be measured once antifungals reach steady-state Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Park WB, et al. Clin Infect Dis. 2012;55:1080; Myers E, Dodds Patterson TF, et al. Clin Infect Dis. 2016;63:e1-60; Park WB, et al. Clin Infect Dis. 2012;55:1080; Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2:55. Ashley E. Curr Clin Micro Rpt. 2015;2: Typically, you want to monitor these drugs once they reach steady-state, which is roughly a week but as soon as four days in some patients. Your goal trough concentration for prophylaxis is typically around 0.5, maybe up to 0.7 for voriconazole. And in response to our audience response question, the trough for treatment of invasive fungal infections is at least 1 for all of these agents. But the take-home message here is, we monitor trough concentrations once the drug has reached steady-state. TDM: How to Incorporate Results? High and low trough concentrations can can be be used to to adjust doses Low Low trough = dose to to meet goals High High trough = dose? (patient dependent) Identify patients who require alternative therapies Unable to to achieve adequate levels with with posa suspension Genetic polymorphisms with with voriconazole De-escalation from combination therapy Combinations often used until until azoles reach therapeutic targets Typically, if levels are low, you increase the dose; if levels are high, you decrease to prevent toxicity. Therapeutic drug monitoring is also important for identifying patients who might require alternative therapies. There are genetic polymorphisms among patients that require them to metabolize voriconazole differently. So, maybe therapeutic drug monitoring helps us identify those patients as well. Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2:55. Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2: The barriers to TDM I think the most important TDM: Barriers barrier is that these assays are not regularly available at all of our hospitals. So, the idea of Availability of a sensitive, timely assay sending out a drug level may take a number of Role of TDM for long-term prophylaxis days at some of your facilities, and that does Voriconazole prophylaxis among lung transplant recipients throw out certain barriers. There s roles for TDM in long-term prophylaxis, but we don t know Cost what to do with some of those levels. And then In In general, don t check a level unless you intend to to do something with the data! finally, I ll just leave you with this: in general, if Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2:55. you re not going to do something with the level, Myers E, Dodds Ashley E. Curr Clin Micro Rpt. 2015;2: don t check it. Oftentimes we get this phone call: I have a level of X; I don t know what to do. And they say, well, why did you check a level? And they don t know. So, don t check a level unless you intend to do something with it. 45

47 Summary Antifungal susceptibility testing provides important information to to optimize antifungal therapy Recommended for for triazoles against Candida blood isolates Encouraged among patients with prior echinocandin exposure Not used routinely for for Aspergillus; case-by-case basis Therapeutic drug monitoring is is recommended for itraand voriconazole (and posaconazole suspension) Measure trough levels at at steady-state to to achieve therapeutic targets In summary, antifungal susceptibility testing is very important for optimizing antifungal therapy. We ve talked about some of the roles for antifungal susceptibility testing. And similarly, I think we have an important role as clinical pharmacists to use therapeutic drug monitoring to optimize our use of these antifungal agents. With that, I ll turn it over to Dr. Slain for a few audience questions and some case scenarios Case Scenarios & Panel Discussion Jointly provided by ProCE, Inc. and the Society of Infectious Diseases Pharmacists (SIDP), and supported by an educational grant from Astellas Scientific and Medical Affairs, Inc. 1 Case 1 A 34-year-old man with a PMH significant only for hypertension, S. aureus endocarditis last year, and prior injection drug use who is admitted to the cardiac surgery service for further evaluation for prosthetic valve endocarditis. He was diagnosed with infective endocarditis of the aortic valve with blood cultures of Candida parapsilosis in multiple blood cultures. Labs Platelets: 52 BUN: 40 ALT: 27 AST: 32 Creatinine: 2.4 Alkaline phosphatase: I made two cases that come from patients that I have seen on my service recently, and my two fellow presenters here are good sports, because I didn t give them details about these cases in advance. I m going to poll you, see what you have to say, and then we ll turn to our experts here on the panel. These are cases that don t have a lot of comprehensive information. And there may be more than one right answer. This is a 34-year-old man with past medical history significant only for hypertension, Staph aureus endocarditis the previous year, and he s a prior injectable drug user. He s admitted to the cardiac surgery service for further evaluation for prosthetic valve endocarditis. Obviously, he had to exchange it with his last bout of endocarditis. He is diagnosed with infective endocarditis of his aortic valve, and he has blood cultures positive for Candida parapsilosis in multiple blood cultures. And you can see that we ve reported a number of labs there. Platelets are 52. You can see the BUN; creatinine; and also, I ve included some LFTs. You can notice that some of those are not normal lab values. Based on this information, which of these regimens would you attempt to go with in this patient? You can see you have a number of options there. Let s poll and see what we get. 46

48 We get a little bit all over the place there. Panel: Would you like to comment as to which option you might pick, given that there s limited information, and you re allowed to, of course, expand if you had certain other diagnostic pieces? Ryan Shields: This is an interesting case. Candida endocarditis is certainly a rare disease, but also one of those severe diseases. And I think when it comes to severe fungal infections, the first thing that crosses my mind is, is there enough benefit of using amphotericin B that outweighs the risks? Amphotericin B has really been the gold standard for endocarditis for a while, and I think that s challenged most recently by new data with the echinocandins, suggesting that the echinocandins are also efficacious. So, when I m looking at this case, I m really debating in my mind between liposomal amphotericin B and a high-dose echinocandin. And the high-dose echinocandin is particularly important here because we have a Candida parapsilosis, which we know have higher MICs to the echinocandins, so there may be some discussion about, is an echinocandin really the best for this? The other thing that I think is important about the pathogen here is you have Candida parapsilosis, which we know can cause biofilms. Biofilms are particularly important in endocarditis, and we know amphotericin B and the echinocandins are usually effective against biofilms; the azoles, less so. So, in my mind I m weighing between amphotericin B and a high-dose echinocandin. So, answers A and E. And I see the patient has an elevated serum creatinine, so it may make you a little bit wary of liposomal amphotericin B; but also, the Candida parapsilosis may make you a little wary of using an echinocandin. So, those are my two choices that I m debating between. I think, for me, it depends on the severity of illness, how aggressive I would be here. Again, in a small case, there s limited things. I think if the patient was really sick, I d probably start amphotericin B and then try to get them to an echinocandin as soon as I could. Melissa Johnson: I tend to agree. The IDSA guidelines here actually say liposomal amphotericin B plus 5- FC is first as combination therapy, or high-dose echinocandin if you can t use amphotericin. I wondered if you were trying to suck us into the fluconazole with the Candida parapsilosis. And I could see some attraction to that, given the renal issues that the patient s experiencing, as well as the concerns about MICs with parapsilosis and candins. But I think a static agent like fluconazole would be ill-advised here, and we should try to go with a cidal agent if we can. Hard-core, I would favor amphotericin B plus 5-FC, but looking at that creatinine of 2.4, you would have to say, where s the patient on the continuum of that? Is that 2.4 and rising, or is that 2.4 and stable? You know, could you manage it or not? And certainly, there are data for high-dose echinocandins here. So, you could maybe do that temporarily while you sort things out with the patient. I was glad to see he was admitted to the surgery service, because the highest recommendation is to consider surgical removal of the valve. Douglas Slain: Those are two great answers. Surgery is really curative, along with the antifungal agent. What we actually did was initiate the high dose of caspofungin. But what we noticed is those LFTs and those liver toxicity indicators started to increase, so we backed down on the caspofungin. We stabilized 47

49 the patient eventually and then the surgeons took over from there. So, that s what happened there. Great answers, though. Susceptibility Report Is Back Fluconazole MIC = 2mcg/mL Caspofungin MIC = 0.25mcg/mL Voriconazole MIC = 0.12mcg/mL Now what would you pick? Here is some additional information. It turns out this is what the MICs were. But based on these MICs, would you change what you would pick? Now, all of these are technically within the susceptibility breakpoint range. Two and less is for fluconazole and caspofungin for this, and 0.12 would be the breakpoint, or less, for voriconazole. So, would it change your vote? 4 Just what we did, doesn t mean it was right, so I m curious to see if anyone changed their vote based on that. So, what do you think, panel? The audience is kind of all over the board too. Would you have stuck with that caspofungin here? Yes. I think a caspofungin of 0.25 is pretty good for Candida parapsilosis. Typically, these cluster around 1 and 2, and that s due to a kind of natural FKS mutation in Candida parapsilosis. So, the MIC looks good, and I think only strengthens the encouragement for an echinocandin to me. I agree. And I think it is nice, though, that you do have a susceptible strain with fluconazole since, eventually, these patients often have to transition to something orally for long, long courses of therapy, and that would at least open up the door for fluconazole as well, if you have to do that in the long term. 48

50 Case 2 A 54-year-old female with AML develops a fever while neutropenic (12 days). She received posaconazole prophylaxis. Her fever has not abated despite receiving piperacillin-tazobactam and vancomycin. Labs Serum creatinine: 1.2 mg/dl LFTs: WNL Presumed diagnosis: Probable pulmonary aspergillosis, based galactomannan and nodules on chest CT 6 How about the second case a 54-year-old female with AML develops a fever while neutropenic at day 12. She receives posaconazole prophylaxis. Her fever has not abated despite receiving piperacillintazobactam and vancomycin. You can see the lab value of the serum creatinine. LFTs are within normal limits. The team has given a presumed diagnosis of probable pulmonary aspergillosis based on galactomannan and nodules on chest CT. This is something we commonly see, especially if you re in a center that is a bone marrow center or treats a lot of AML patients. So, these are your options. Give it a vote here and tell us what you might pick for a patient like this. A little bit of spread here. Panel, would you like to address maybe what approach you would take? Again, you have limited information, of course. Melissa Johnson: Yes, I agree. Douglas Slain: But you know what agent was used for prophylaxis. Melissa Johnson: With this information, I ve got a million things running through my head and I would ask about 80 questions of the team before deciding. Certainly, it s concerning that she was on posaconazole prophylaxis when this happened, and raises the question of, what formulation was she on; did she have therapeutic levels; did she have malabsorption? Is this really Aspergillus or is it something else? And because of that question, we often start broadly with lipid amphotericin B until we really have histopath and sort things out. But if it is Aspergillus, you ve got to start with voriconazole, based on the data. So, I think those would be the two I d be thinking about. Isavuconazole, I might consider, if she has some QT prolongation issues or we re worried about toxicities. That would also be in the back of my mind too, because you do have broader mold coverage with isavuconazole as well. Ryan Shields: Yes, I agree. You have a breakthrough case here and somebody on azole prophylaxis, so I think I d be inclined to start with lipid amphotericin B. 49