The current treatment landscape: other fungal diseases (cryptococcosis, fusariosis and mucormycosis)

Abstract

Compared with major invasive mycoses such as aspergillosis and candidiasis, the antifungal stewardship management strategies of other fungal diseases have different opportunities and considerations. Cryptococcosis, fusariosis and mucormycosis are globally prevalent invasive fungal diseases (IFDs), but are not currently included in antifungal prophylaxis guidelines for immunocompromised hosts. Since the implementation of biomarkers as part of diagnostic screening strategies, the concept of pre-emptive antifungal therapy has emerged for these IFDs. Management of cryptococcosis, the most common IFD worldwide, generally utilizes a pre-emptive or therapeutic strategy that does not involve prophylaxis or empirical antifungal treatment strategies. Antifungal stewardship outcomes for cryptococcosis may vary according to the availability of local resources. Invasive fusariosis, the second-most common form of non-Aspergillus mould infection among haematological malignancy patients, can be managed with pre-emptive (or diagnostic-driven) approaches based on the monitoring of serum galactomannan (GM) antigen in increased-risk populations. The success of antimicrobial stewardship programmes in decreasing the burden of invasive fusariosis in selected patient populations depends on the development and implementation of rapid diagnostic strategies for early and appropriate administration of therapy. Mucormycosis may emerge as a breakthrough IFD in haematology or solid organ transplant recipients receiving antifungals that lack activity against Mucorales. The concept of pre-emptive antifungal therapy has thus arisen for mucormycosis in the haematology setting because of the recent availability of circulating Mucorales DNA measurement. These examples demonstrate the challenges of implementing antifungal stewardship programmes in areas with limited resources, as well as in IFDs that are difficult to diagnose and treat.

Introduction

In this article we consider invasive fungal diseases (IFDs) other than aspergillosis and candidiasis and their impact on antifungal stewardship (AFS) management strategies. Firstly, cryptococcosis will illustrate that AFS may vary according to local resources (i.e. the availability of antifungal agents). Next, fusariosis and mucormycosis will be discussed as two difficult-to-treat IFDs, particularly in the haematology setting. These three IFDs are not currently included in antifungal prophylaxis guidelines in immunocompromised hosts, and until recently the antifungal management decision was driven by strict mycological or histological documentation (i.e. probable or proven IFD only). Since the implementation of biomarkers as part of diagnostic screening strategies, the concept of pre-emptive antifungal therapy has emerged for cryptococcosis among HIV-infected patients in highly prevalent geographical areas. Even though robust non-culture diagnostic tools have not yet been developed for the timely and accurate diagnosis of fusariosis and mucormycosis in the haematology setting, these are strong examples of diseases where judicious use of appropriate antifungal therapy is paramount. These two fungal diseases are therefore also discussed in this article.

Cryptococcosis

Cryptococcosis represents the most common IFD worldwide. Before widespread antiretroviral therapy (ART) was available, estimated cases per year reached ∼1 million,¹ and even with improved HIV treatments the number of cases and mortality rate remained substantial. For AFS in the management of cryptococcosis, the message may be slightly different than for other major invasive mycoses such as aspergillosis and candidiasis. There are several unique aspects to cryptococcosis that have a significant impact on how antifungal agents are utilized and how stewardship advice is provided to the medical community. First, there have been no guideline recommendations to use antifungal agents as prophylaxis against cryptococcosis in high-risk patients.² The highest identified risk population worldwide for cryptococcosis is the HIV-infected individual with CD4 cell counts under 100 cells/mm³. Furthermore, early in the HIV epidemic and before widespread use of potent ART, it was shown that the prophylactic use of fluconazole in HIV-infected individuals with low CD4 counts could actually prevent the appearance of cryptococcosis.³ However, fluconazole treatment as a prophylactic strategy was not recommended for several reasons: (i) it had not been shown that fluconazole treatment had an impact on survival; (ii) ART evolved to allow the risk period to be relatively short; (iii) before widespread prophylactic use of fluconazole, one would need an assessment of the general prevalence of cryptococcosis in the specific geographical population for its use, in order to show potential clinical impact; (iv) there were significant concerns that the widespread use of a fungistatic antifungal such as fluconazole, in a large number of patients, could select for drug-resistant yeasts in the host population; (v) the cryptococcal polysaccharide antigen (CRAG) test is a relatively accurate and simple diagnostic test that can identify early disease. It is the CRAG test that has controlled the use of early antifungal agents in cryptococcosis. This test demonstrates the effectiveness of integrating clinical risk assessment with an appropriate and validated biomarker for reducing the use of antifungal agents in clinically identified high-risk patients.

Although primary prophylaxis for cryptococcosis is unlikely to be useful, the next issue for AFS focuses on the optimal use of the CRAG biomarker. With the development of a simple, cheap, sensitive and accurate lateral flow assay for detection of the cryptococcal polysaccharide antigen,^4,5 management strategies in high-risk patients in certain geographical locations could efficiently use a CRAG screening test in ‘at-risk’ patients (<100 CD4 cells/mm³ and HIV infection) to intervene at an earlier stage of infection and/or guide further diagnostic work-up, such as a lumbar puncture.⁶ The CRAG test is able to detect subclinical cryptococcal disease in patients entering ART programmes. When a CRAG-positive patient is identified, fluconazole pre-emptive treatment can prevent many of these patients from developing severe cryptococcosis. Furthermore, CRAG identifies a high-risk patient at risk of a poor outcome.⁷

Following the advent of CRAG screening and a pre-emptive fluconazole strategy, there have been a series of cost-effective analyses for this approach^6,8–13 and an endorsement in the WHO guidelines (‘Rapid advice: diagnosis, prevention and management of cryptococcal disease in HIV-infected adults, adolescents and children’) for the strategy.¹⁴ The strategy has now been adopted by some high-prevalence countries, such as South Africa and Uganda.¹⁵ Furthermore, there have been estimates of the costs for pre-emptive therapeutic strategies with CRAG screen testing and fluconazole administration. It is important to note that CRAG tests are now cheap and that fluconazole as an oral regimen is relatively inexpensive. However, some less-resourced countries may still find these costs substantial within their healthcare resource budgets. With CRAG-negative patients, clinicians will have substantial assurance that fluconazole treatment is not needed and cryptococcosis will not generally be a problem for the patient. On the other hand, CRAG-positive patients have substantial management issues and exemplify high-risk patients with substantial failure rates. The ability to set up CRAG screening of high-risk patients in high prevalence areas such as sub-Saharan Africa is continuing to be studied and/or implemented in HIV clinics.^6,15 Healthcare systems need to understand their prevalence of disease, with careful epidemiological studies to predict the cost-effectiveness. For instance, it is likely that as prevalence of CRAG positivity in the risk population reaches 3%, a CRAG testing system and early management of asymptomatic and symptomatic patients with fluconazole or amphotericin B, respectively, may allow early, effective antifungal therapy in a very focused manner. In many respects, this pre-emptive strategy for the early use of antifungal therapy directed to at-risk hosts represents an important paradigm for all AFS principles in the era of frequent antifungal drug empiricism. It eliminates widespread use of antifungal prophylaxis to reduce costs and the potential risk of drug resistance development, and integrates biomarker use in a clinical strategy to attempt early intervention with antifungal therapy.

The next level of AFS for cryptococcosis revolves around induction therapy for cryptococcal meningoencephalitis. In this clinical situation, stewardship models for cryptococcosis need to be divided into ‘resource-available’ and ‘resource-limited’ groups. First, polyene use will be highlighted. In resource-available countries there has been a gradual shift from amphotericin B deoxycholate to the lipid formulation of amphotericin B (LFAB) due to substantial decreases in toxicity.¹⁶ It has been critically observed how important uninterrupted induction therapy is to treatment success in cryptococcal meningoencephalitis, and this is much more likely to occur with LFAB than standard amphotericin B.¹⁷ Therefore, even without direct comparative combination studies with LFAB, clinical practice has shifted towards the use of LFAB as a combination therapy with flucytosine (5-fluorocytosine) despite its higher acquisition costs. In resource-limited countries, the primary focus has simply been on induction-phase delivery of amphotericin B deoxycholate for a fungicidal induction therapy for at least 1–2 weeks. With its toxicity issues, intravenous administration and costs, amphotericin B deoxycholate treatment has not been an easy therapeutic option for many of these countries and an inferior induction regimen of fluconazole is frequently used.

The primary recommended induction therapy for cryptococcal meningoencephalitis is a polyene plus flucytosine.² However, flucytosine remains a therapeutic drug enigma. In many areas of the world it is simply not available, since many countries do not have this drug licensed. On the other hand, in resource-available countries such as the USA, there has been a flucytosine acquisition price increase, which has now created a $23 842 cost per quality-adjusted life year for the combination of a polyene and flucytosine in cryptococcal meningoencephalitis.¹⁸ This dramatic increase in flucytosine costs to ∼$2000 per day for 2 weeks now matches the entire hospitalization costs for cryptococcal meningoencephalitis treatment only a decade ago.¹⁹ Therefore, flucytosine represents a bilateral drug issue in that it is becoming too expensive for resource-available countries, and it is not available in resource-limited countries. For the ideal management of this infection, which still has a 20%–50% treated mortality, it is imperative that there are new initiatives to provide more balance in the acquisition of these life-saving drugs.

There are also stewardship caveats. If flucytosine is not available, then should treatment with amphotericin B alone or amphotericin B plus fluconazole be administered? Studies are probably more favourable to the use of the combination of a polyene and azole for induction therapy.²⁰ In cryptococcal meningoencephalitis, the strategy has consisted of three parts: (i) induction; (ii) consolidation; and (iii) suppression. This was first studied in AIDS patients before the establishment of ART.²¹ The strategy studied in AIDS patients has been adopted for all patient groups with disseminated cryptococcal disease, but without careful study. It is clear that in cryptococcosis, like many IFDs, AFS would be significantly aided by more robust, prospectively designed studies to determine optimization of specific drug regimens and doses and more precise lengths of antifungal therapies.

In summary, cryptococcosis management generally utilizes a pre-emptive or therapeutic strategy and does not involve prophylaxis or empirical antifungal treatment strategies. However, further understanding of the powerful CRAG-derived, pre-emptive strategy needs more validation of the impact and costs within identified high-risk patient groups. The development of new CRAG-detection tools, including a semi-quantitative measurement, might be valuable for the early discrimination of patients with latent cryptococcal meningitis. These tests, such as the one developed at the Institut Pasteur in partnership with Biosynex (Immunoquick CryptoPS),²² are currently being evaluated in the field. In the therapeutic arena for a proven invasive disease, there are some fixed principles for the values of fungicidal regimens and combination therapy,¹⁶ but practically there remain substantial road blocks for availability of drugs and this compromises successful outcomes.

Fusariosis

Although the genus Fusarium comprises at least 70 different species, only 12 of them have been found to cause clinical disease in humans, mostly in the form of superficial infections such as onychomycosis or fungal keratitis in immunocompetent individuals.²³ In contrast, invasive fusariosis poses a life-threatening complication among severely immunocompromised hosts, with case fatality rates reaching 70% for disseminated forms.²⁴ Most episodes of invasive fusariosis are due to two species complexes: F. solani complex (accounting for ∼50% of episodes) and F. oxysporum complex (accounting for 20%). The remaining cases are mostly due to F. incarnatum–F. equiseti species complex, Gibberella fujikuroi species complex and F. dimerum species complex.²³

Despite the ubiquitous distribution of the Fusarium species and their saprophytic nature in plants, animals and humans, invasive fusariosis represents an emerging condition with an increasing number of case reports and series over the last few years.^25,26 In fact, invasive fusariosis ranks as the second-most common form of mould infection among haematological malignancy patients in some recent series, surpassed only by invasive aspergillosis.²⁷ Several factors may contribute to this rising epidemiology, including improvements in diagnostic methods. Innate immunity plays an instrumental role in controlling angioinvasion and tissue destruction by Fusarium species. Thus, prolonged neutropenia (such as that associated with remission-induction therapy in patients with acute myeloid leukaemia) is consistently identified as a major risk factor for invasive fusariosis, as well as a marker for poor prognosis.^24,28 In addition, environmental factors (contamination of water reservoirs, for instance) have been involved in the epidemiology of hospital outbreaks of fusariosis.^29,30 Finally, although no clinical breakpoints have been formally defined to date, most Fusarium species exhibit decreased in vitro susceptibility to a large number of currently available antifungal drugs,²³ with amphotericin B usually being the agent with the lowest MICs.²³ This points to the potential impact of selective pressure due to the widespread use of azoles and echinocandins on the risk of IFD due to this mould pathogen.

A number of studies have assessed the incidence and microbiological features of breakthrough IFD in patients receiving echinocandins.^31–33 Although infrequent, this complication usually involves a poor prognosis, with 12 week survival rates ranging from 33%³⁰ to 42%.³¹ Invasive fusariosis is common in such a scenario,³⁴ contributing to one-quarter of cases of proven breakthrough IFD in some series.³² On the other hand, the occurrence of invasive fusariosis has also been well documented during the administration of broad-spectrum triazoles, with various cases reported in the literature.^35–39 In a retrospective review from a single cancer centre comprising 44 cases of proven or probable invasive fusariosis, breakthrough infection was demonstrated in as many as one-third of patients, with most of them receiving prophylaxis with voriconazole or posaconazole.²⁴ A recent literature review on disseminated fusariosis found that half of the patients were on systemic antifungal therapy at the time of diagnosis.³⁹ Despite the lack of specific studies, the above-mentioned studies would suggest the potential role of AFS programmes in decreasing the burden of invasive fusariosis in selected patient populations.

The process of antimicrobial stewardship is also founded on the early and appropriate administration of therapy, raising particular interest in the implementation of rapid diagnostic strategies.⁴⁰ Although the Platelia Aspergillus serum galactomannan (GM) antigen assay is usually considered specific for invasive aspergillosis, it has been demonstrated to possess cross-reactivity with Fusarium species.^41–43 This fact may be used to design pre-emptive (or diagnostic-driven) approaches against invasive fusariosis based on the monitoring of serum GM antigen in increased-risk populations, particularly in settings with high prevalence of invasive fusariosis. A small retrospective study carried out in Brazil among 18 haematological malignancy patients (most of them with IFD due to F. solani species complex) reported sensitivity and specificity values of a positive serum GM (cut-off value of 0.5) of 83% and 67%, respectively. Of note, serum GM preceded the diagnosis of fusariosis by a median of 10 days (range 3–39 days).⁴⁴ Fungaemia is a distinctive feature of invasive fusariosis as compared with other forms of opportunistic mould infections. The prompt identification of isolates to the species or species-complex level would allow the rapid initiation of therapy with drugs that have been shown to be more effective against Fusarium species (voriconazole,⁴⁵ amphotericin B or posaconazole), although the clinical response rates are modest at best. The application of MALDI-TOF MS in the clinical setting has improved the identification of bacteria and fungi. A recent study evaluated the performance of the MALDI-TOF MS assay to discriminate between species complexes and specific species in 289 Fusarium strains, reporting no incorrect species complex identifications. In addition, >80% of the identifications were correct to the species level.⁴⁶ These preliminary experiences may eventually offer an opportunity for implementing AFS programmes with the ability to impact the incidence and outcome of invasive fusariosis.

Mucormycosis

Mucormycosis is a major IFD caused by filamentous fungi belonging to the Mucoromycotina subphylum with broad, thin-walled, sparsely septate, ribbon-like hyphae. The latter are ubiquitous (found in hospitals and homes) and saprophytic (found in soil, decaying vegetation and fruits), with the genus and species being influenced by geography. These moulds can sporulate and air-borne dispersed spores may colonize and infect sinuses, as well as the lower respiratory tract, and less frequently the gastrointestinal tract and skin.⁴⁷ Ten genera cause human disease but there are two predominant genera in the Western world: Rhizopus and Lichtheimia.⁴⁷ Innate immune cells such as alveolar macrophages, monocytes and neutrophils are key players in fungal recognition. To further increase subsequent therapeutic complexity, it should be noted that Mucorales also interact with endothelial cells, explaining why mucormycosis is a true fungal vasculitis with subsequent thrombosis and necrosis. Both acidosis, which increases the concentration of free iron available for fungal growth, and hyperglycaemia, which reduces fungal phagocytosis, are critical for mucormycosis pathogenesis.⁴⁸ The incidence of mucormycosis has been estimated in France to be 0.09/100 000, representing 1.5% of 35 876 cases of IFD, with an increase of 7.3% per year and an increasing death rate.⁴⁹ In addition, Mucorales are increasingly responsible for healthcare-associated outbreaks,^50,51 and haematological malignancy patients now represent 50%–77% of mucormycosis cases; those with diabetes mellitus represent ∼20% of cases at least in Europe and the USA.^52,53

There is also a third emerging group of those who present with mucormycosis acquired post-trauma, which represents ∼20% of the cases in recent series.^52,53 The traumas in these cases include natural disasters such as tornados⁵⁴ and combat-related wounds.

Due to the rarity of the disease, there is no primary antifungal prophylaxis against mucormycosis recommended, even in high-risk groups. However, mucormycosis may emerge as a breakthrough IFD in haematology or solid organ transplant recipients receiving antifungals that lack activity against Mucorales, such as voriconazole and the echinocandins.^55,56

Mucormycosis diagnosis suffers from a lack of antigen detection, either β-glucans or specific biomarkers, as well as a lack of antibody detection systems. Thus, until recently, and still in most laboratories, reference diagnostic methods remain as microscopy and culture, the latter benefiting from the recent availability of molecular tools for the identification of genus/species in culture and tissues.⁴⁷ Two recent major advances may change the therapeutic approach for mucormycosis, particularly among haematological malignancy patients, which represent the largest high-risk group, and may support the concept of pre-emptive treatment of the disease. The first is the use of chest CT for an early diagnosis of pulmonary lesions, with the presence of the reverse halo sign in >90% of the cases at the early stage, although this disappears later on, particularly after 15 days.⁵⁷ The second is the combination of a triad of quantitative PCR assays targeting 18S rDNA of Mucor species/Rhizopus species, Lichtheimia complex and Rhizomucor species in serum, which allows earlier diagnosis up to 2 months before the final diagnosis.⁵⁸

Although there are no breakpoints defined and no clear in vitro/in vivo correlation evidence for antifungal susceptibility testing of Mucorales, acceptable MIC values are found for polyenes, except for Cunninghamella species, and for only two triazoles, posaconazole and isavuconazole.^59,60 However, it should be noted that the range of MIC values for isavuconazole is large⁶⁰ and whether or not the latter agent is as effective when targeting Mucorales harbouring high MIC values should be further investigated.

From the therapeutic point of view, the results of three prospective clinical trials (in combination with surgery when needed) have been available over the past 5 years. One open multicentre study, AmBizygo, tested the clinical value of high-dose liposomal amphotericin B (10 mg/kg/day) as a first-line treatment of mucormycosis. Response was analysed in 33 patients and most patients had a haematological malignancy as their primary underlying disease (53%). At week 12, there was a 45% favourable response rate and 38% mortality rate.⁶¹ A second trial, named Defeat Mucor, tested liposomal amphotericin B and deferasirox, a powerful iron chelator, in a multicentre, randomized, double-blind study with liposomal amphotericin B versus placebo (liposomal amphotericin B at ≥5 mg/kg/day and deferasirox at 20 mg/kg/day for 14 days). The results were disappointing as both response rate and survival were poorer for the tested combination.⁶² The most recent trial, the Vital study, investigated the efficacy of isavuconazole (median duration of 84 days) in 46 patients with mucormycosis, including 21 patients as a first-line therapy. At the end of treatment, the success rate was 31.6% and mortality rate was 42.9% at day 84.⁶³ The latter results led to the recent registration of isavuconazole for the treatment of mucormycosis in Europe and the USA.

Finally, two European therapeutic guidelines for the management of mucormycosis, one from the European Conference on Infections in Leukaemia (ECIL) and the other from ESCMID and the European Confederation of Medical Mycology (ECMM) study groups, were recently published. As first-line therapy and before the availability of isavuconazole, they both recommend the use of a lipid formulation of amphotericin B, with the liposomal formulation at 5 and 10 mg/kg/day in cases of brain involvement, in combination with radical surgery when feasible.^64,65 No recommendation for combined antifungal therapy was mentioned as a first-line strategy, and because of the lack of data on posaconazole as a primary treatment, the latter drug was only considered as a second-line strategy.

In conclusion, over the past 2 years there has been a dramatic change in the management of mucormycosis thanks to the availability of a pertinent biomarker, namely quantitative PCR, which now allows improved targeted antifungal therapy to be used.

Summary

There are key challenges associated with antifungal stewardship management strategies for cryptococcosis, fusariosis and mucormycosis, which differ from those faced for other fungal diseases, such as candidiasis and aspergillosis. It is important to address these challenges with appropriate strategies in order to optimize treatment for patients with these diseases. Cryptococcosis provides an example of the importance of considering the availability of local resources as part of the stewardship programme, and highlights how AFS has the potential to reduce the widespread use of unnecessary antifungal prophylaxis to reduce costs and the potential risk of drug resistance development. Fusariosis and mucormycosis also provide an example of the potential role of AFS programmes in decreasing the burden of disease in selected patient populations.

Transparency declarations

M. F.-R. holds a clinical research contract ‘Juan Rodés’ (JR14/00036) from the Spanish Ministry of Economy and Competitiveness (Instituto de Salud Carlos III). O. L. has been a consultant for Gilead and Novartis, speaker for Gilead, Novartis, Merck, Pfizer and Astellas. J. R. P. attends Advisory Committees for Astellas, Merck, F2G, Cidara, Scynexis, Viamet, Arno, Vical and Amplyx.

This article forms part of a Supplement sponsored and funded by Gilead Sciences Europe Ltd; editorial assistance was provided by Synergy Medical. The content of this Supplement is based on the sessions presented at the CARE VIII meeting, held in Madrid in November 2015.

References