Antifungal stewardship in clinical haematology

Antimicrobial stewardship has overwhelmingly focused on antibiotics but antifungal stewardship is now just as crucial. Invasive fungal disease (IFD) has a lower hospital incidence in comparison to infections caused by multi-resistant bacteria, but its health and financial burden are substantial. The incidence of IFD is increasing in the Western countries each year as a result of improving medicine and an increase in survival rates from previously fatal diseases.(1) The emergence of a wider selection of immunosuppressive agents, especially those used in haemato-oncology patients, significantly increases the risk of developing a serious fungal infection.

IFD is common in high-risk patients with haematologic malignancies, such as patients with acute leukaemia receiving induction chemotherapy, and cause substantial morbidity and mortality. This remains a major concern for clinicians managing this patient population. The risk for invasive fungal infections increases with the duration and severity of neutropenia, prolonged antimicrobial use and number of chemotherapy cycles.(1)

The high drug costs and toxicities of antifungal agents are the principal rationale for antifungal stewardship, while antifungal resistance is an emerging but less prevalent issue. There have been suboptimal diagnostic tools, which have driven the overuse of antifungal agents. One of the most challenging parts of antifungal stewardship to implement is de-escalation of empirical treatment. Incorporating non-culture-based tests into clinical pathways may enhance antifungal stewardship.(2)

Antifungal stewardship efforts can improve the appropriate and effective use of antifungal agents. Recognising the risk factors can guide the optimal use of antifungal prophylaxis and treatment for at risk patients.

This patient population is at risk of invasive fungal infections caused by both yeasts and moulds. A yeast is any of the various unicellular fungi that reproduce by budding and form ascospores and are capable of fermenting carbohydrates. This includes Cryptococcus and Candida spp. A mould is a multicellular organism with filamentous hyphae, including the Aspergillus spp, that reproduce either asexually or sexually by spore formation. Moulds obtain their nutrients from the organic sources.

Candida albicans has been the leading cause of yeast infections in haematopoietic stem cell transplant (HSCT) recipients, but the current widespread use of azole prophylaxis in the early transplant period has led to the emergence of a variety of non-albicans species, such as C. tropicalis, C. krusei and C. glabrata, as important pathogens. These yeasts are normal inhabitants of the skin and the oral and gastrointestinal mucosa. Breakdown of these mucosal surfaces due to radiation and chemotherapy, compounded by neutropenia in the pre-engraftment period, can greatly increase the risk of invasion and systemic infections. Presence of in-dwelling central venous catheters and alteration of  normal surface flora due to antibiotics

are additional risk factors for Candida infections.(1)

Most mould infections in HSCT recipients are due to A. fumigatus, A. flavus, and A. niger, which gain entry through breakdown of skin or mucosal surfaces or through the nasal passages and respiratory tract. Aspergillus infections can occur early after HSCT (during the neutropenic phase) or later, especially complicating the immunosuppression associated with acute or chronic graft-versus-host disease (GvHD). Risk factors for Aspergillus infections include allogeneic (more than autologous) HSCT, prolonged neutropenia, and GvHD. Transplantation for certain diseases, such as chronic granulomatous disease, aplastic anaemia, and myelodysplastic syndrome, is associated with a higher than usual risk of aspergillosis. History of Aspergillus infection has been observed to be a risk factor for reactivation after HSCT.(1)

Table 1 provides a guide to the clinical susceptibility of fungi.(3)

Recommendations for prophylaxis, diagnosis and treatment of invasive fungal infections, whether empirical or definitive, vary across guidelines. Currently the majority of guidelines including the British Committee for Standards in Haematology guidelines (BCSH) are outdated and revised versions are eargely awaited.

Determining the most appropriate antifungal prophylaxis regimen, whether primary or secondary, is challenging for several reasons. Many guidelines suggest risk stratifying patients from very low risk to very high risk of developing IFD, to determine the most appropriate antifungal regimen. Unfortunately, there is no uniformity across the guidelines for the criteria for each risk group and, in some cases, such as BSCH guidelines, a simple list is used to identify those patients at higher risk. These criteria include those with GvHD, neutropenia for >seven days, long-term corticosteroids, bacterial infections and mucosal colonisation which are all risk factors for both Candida and Aspergillus infections. Cytomegalovirus (CMV) infection and exposure to building works are listed as risk factors for infection with Aspergillus. Antibacterials and a central venous catheter are risk factors for infection with Candida.(4)

Primary antifungal prophylaxis (PAP) involves administration of antifungal chemotherapy to patients who have no evidence of infection but whose epidemiological risk profile indicates a high propensity for the development of invasive fungal infection. The benefits of primary antifungal prophylaxis have been observed in multiple studies. Much of the emphasis historically has been on the prevention of Candida infections. However, studies of aspergillus prevention have intensified in recent years.

Azoles are recommended for prophylaxis in high-risk patients.(4–8) European Conference on Infections in Leukaemia-3 (ECIL-3) and German Society for Haematology and Medical Oncology guidelines recommend that the choice is based on clinical indication. For example, during the neutropenic phase following an allogeneic stem cell transplant, fluconazole is recommended, but this is to be switched to posaconazole if the patient were to develop GvHD. If fluconazole is chosen, there is a need for a mould-directed diagnostic strategy. Two recent voriconazole studies show no survival benefit or reduction in invasive aspergillosis in patients given voriconazole prophylaxis, as compared with the control arms of itraconazole or fluconazole.(9,10)

Nevertheless, the ECIL-3 guidelines recommend voriconazole for both the neutropenic and GvHD phases following allogeneic stem cell transplantation. For patients with acute myeloid leukaemia or myelodysplastic syndrome receiving intensive chemotherapy, posaconazole is recommended.(11) Although given the variability in drug levels observed with itraconazole therapeutic drug monitoring (TDM; see below), it might be appropriate to optimise dosing and avoid breakthrough infections. This strategy should allow an objective choice of antifungal treatment with suspected or proven IFD thereafter.

Secondary anti-fungal prophylaxis (SAP) involves the administration of antifungal therapy to a patient who is undergoing a period of immunosuppression and who has a history of previously documented and fully resolved invasive fungal infection(s). The efficacy of SAP in patients with active infection or with persistent radiological abnormalities remains unclear. The ECIL-3 guidelines suggest that the choice of drug should be based on the previously isolated

pathogen and previous response to antifungal therapy.(12)

Patients with prior invasive fungal infection, especially Aspergillus infections, are at high risk for recurrence of infection with further chemotherapy. This has mainly been studied in patients with prior Aspergillus infection. It can prevent reactivation of infection in most patients and permit further chemotherapy. For practical purposes, SAP can be regarded as treatment rather than prophylaxis and therefore requires an agent with activity against proven mould infection.

The choice of agent depends in part upon the need to avoid drug interactions while chemotherapy is being administered and toxicity profiles, particularly in patients with pre-existing organ impairment.

However, as voriconazole is recommended as a first- line agent for proven or probable IFD, this may pose an issue when using other mould active azoles prophylactically in the same population, because of the risk of selection of resistant strains in patients who fail prophylaxis. Few alternative options with similar efficacy and tolerability to voriconazole exist for the treatment of potential breakthrough invasive aspergillosis. Generally, in the case of breakthrough of invasive aspergillosis under azole prophylaxis, it is recommended to switch to another class of mould-active antifungals, preferably liposomal amphotericin or caspofungin.(11)

Primary antifungal prophylaxis during conventional chemotherapy for acute leukaemia should be restricted to the duration of neutropenia. There is no evidence that prolonged therapy in this setting is of any benefit. However, prolonged therapy may be needed in secondary prophylaxis. The duration of anti-fungal prophylaxis should be individualised based on the patient’s clinical status and history of prior fungal infections.

If an IFD is clinically suspected, a biopsy or aspiration of appropriate fluid should be obtained without delay, with the aim of making a diagnosis of proven IFD. Antifungal therapy is justified in cases where microscopic finding or growth of any yeast or mould from any normally sterile tissue or fluid source is observed. Only the BCSH and Australasian Society for Infectious Diseases guidelines provide recommendations for computed tomography (CT) scans.(4,13) This practice is common in the UK. CT findings in intra pulmonary aspergillosis range from single or multiple nodules to mass-like infiltrates and the halo sign. Late signs include cavitation, with or without the air crescent sign.

BCSH recommends that cases where clinical features are consistent with invasive pulmonary aspergillosis (IPA), with or without positive fungal cultures, justifies an urgent chest CT irrespective of X-ray findings. Patients with inconclusive clinical or microbiological evidence of an IFD and negative, non-specific CT findings, should have a repeat CT scan within seven days. Axial and coronal CT of the sinuses and surrounding structures should be undertaken immediately on clinical suspicion of sinus infection.(4)

Galactomannan (GM) testing is now widely accepted as a sensitive method for assessing patients with a haematological malignancy that are at risk of an IFD. GM testing of plasma, cerebrospinal fluid and bronchoalveolar lavage fluid as well as serum is recommended to support the diagnosis of Aspergillus infection.(2)

Polymerase chain reaction (PCR) and GM-enzyme-linked immunosorbent serologic assay (ELISA) for Aspergillus are also effective diagnostic tools in this scenario. Bronchoalveolar lavage fluid may be superior to blood for ELISA and PCR. The use of a combination of clinical, molecular, serological and radiological evidence provides a robust, non-invasive diagnostic tool for the diagnosis of IFDs.(2)

Several factors have positively impacted on the outcomes of these patients, such as use of prophylactic antifungals, local incidence and epidemiology of IFDs, availability of diagnostic tests and high-efficiency particulate absorption filtration.

This is use of antifungals in an immunosuppressed patient with clinical signs and symptoms suggestive of IFD. Symptoms include persisting fever of unknown source despite appropriate empirical antibacterial therapy and in the absence of supporting evidence of invasive fungal infection. This strategy was first introduced as a means to prevent invasive fungal infections in the 1980s after it was noted that many patients with fevers had underlying, otherwise undiagnosed, fungal infections, and particularly invasive Candidiasis.(14)

Pre-emptive antifungal therapy is the use of antifungals in an immunosuppressed patient following confirmatory laboratory or radiological tests, but in the absence of definitive culture or histopathological evidence. Pre-emptive therapy is an early treatment strategy that has been proposed as an alternative to empirical therapy. Pre-emptive therapy involves initiating antifungal therapy based upon the results of serial screening for aspergillosis. Pre-emptive treatment was found to be non-inferior to empirical therapy when assessing survival two weeks after recovery from neutropenia.(11)

Based on clinical experience, many groups suggest that empirical treatment is used for a patient at high risk who has persistent or relapsing fever refractory to broad-spectrum antibiotics, without an identified cause for 72h or more.(14)

The BCSH guidelines recommended liposomal amphotericin B and caspofungin as first line agent where IFD is suspected, the latter having a superior toxicity profile. In proven, or probable, central nervous system (CNS) IFD, voriconazole is the drug of choice because of its superior CNS penetration.(4) Aspergillus-specific recommendations are offered by the IDSA, in 2008, where grade A evidence indicates that voriconazole and liposomal amphotericin can be used.(8)

ECIL-3 guidelines do make some recommendations for combination therapy in special circumstances, although does discourage use as first-line therapy.(11)

Failure of treatment is defined as progression of clinical features (clinical deterioration with worsening symptoms, rising CRP and persistent fever) after at least seven days of adequate antifungal therapy. It is recommended to consider changing to a appropriate second-line agent. The specific choice of agent may vary on an individual patient basis and should ideally be discussed in a multi-disciplinary team setting involving consultant microbiologists, haematologists and senior pharmacists. There are limited data available for salvage treatment.(15)

As our understanding of antifungal pharmacology, including pharmacokinetics and pharmacodynamics, improves, this has led to TDM becoming an important tool that could be utilised when administering some antifungal agents.(16) TDM may assist in improving patient outcome and prevent toxicity and resistance. The recent TDM guidelines for antifungal agents from the British Society for Medical Mycology review the literature in this area and provide a number of recommendations. The review does state that majority of the evidence supporting the TDM is circumstantial and there it is debatable whether TDM of antifungal agents should be routine as for other antibiotics or selective. Antifungal TDM is expensive and time consuming and clinical input and judgement should remain central to the TDM process.

Key recommendations from this guideline are that TDM should be performed for the majority of patients receiving itraconazole in the first week of treatment and regularly thereafter. The evidence for TDM improving clinical benefits in this cohort is strong, based on the considerable pharmacokinetic oral bioavailability affected by food and gastric pH, significant compliance issues and evidence suggesting relevant drug exposure–response and drug exposure–toxicity relationships. Thus, TDM for itraconazole could aid the minimisation of drug-related toxicity. The target trough concentration for prophylaxis and treatment is 0.5mg/l and >0.5mg/l, respectively.

Another key recommendation is that TDM should be performed for the majority of patients receiving voriconazole in the first five days of treatment and regularly thereafter. There is now increasing evidence to support TDM for voriconazole. The target trough concentration for prophylaxis and treatment is >1mg/l. The guideline also recommends TDM for the majority of patients receiving posaconazole in the first week of treatment and regularly thereafter. The target trough concentration for prophylaxis is >0.7mg/l at steady state or 0.35mg/l 48 hours post-initiation and >1mg/l for treatment.

It is important to perform TDM for these antifungal agents when interacting drugs start or stop, or following dose adjustments. TDM can also be useful if the clinician is uncertain about compliance or there are concerns about gastrointestinal absorption, especially for prolonged periods of time or when potential clinical or laboratory manifestations of toxicity occur.

Strategies for dosage adjustments that may be required in patients with low serum concentrations or other measures that may need to be taken are included in this guideline.

Improvements must be made to improve management of invasive fungal infections. Most important is the requirement for a multidisciplinary team to ensure an accurate diagnosis and that appropriate treatment is delivered promptly. This team should include a haemato-oncologist, a radiologist, a microbiologist, a pharmacist and a respiratory physician. Multidisciplinary diagnostic-driven management guidelines/pathways are required locally, which incorporate non-culture-based tests, such as galactomannan and PCR, along with CT. Antifungal stewardship in this area remains challenging.