Abstract
Invasive fungal infections (IFIs) have become high prevalence in patients with hematologic malignancies. Drug-based strategies for IFIs include various approaches such as prophylactic, empiric, preemptive, and directed treatment. Prophylaxis is an attractive strategy in high-risk patients, given the lack of reliable diagnostics and the high mortality rate associated with IFIs. Prophylaxis includes the use of antifungal drugs in all patients at risk. An ideal antifungal compound for prophylaxis should have a potent and broad activity, be available both orally and intravenously, and have a low toxicity profile. Voriconazole fulfills all these criteria. The clinical efficacy of voriconazole against the majority of fungal pathogens makes it potentially very useful for the prevention of IFIs in patients with hematologic malignancies. Voriconazole appears to be very effective for the primary and secondary prevention of IFIs in these patients and recipients of allogeneic hematopoietic stem-cell transplantation. Randomized controlled trials evaluating voriconazole as primary antifungal prophylaxis in patients with neutropenia treated for a variety of hematologic malignancies have been performed, confirming its value as a prophylactic agent. Voriconazole is generally safe and well tolerated; however, its use is also associated with a number of concerns. In most patients with hematologic malignancies there is the potential for pharmacokinetic drug–drug interactions given that voriconazole is metabolized through the P450 cytochrome system.
Introduction
Although there have been major improvements in the past few years in the diagnosis and treatment of invasive fungal infections (IFIs), they remain among the most common causes of infectious morbidity and mortality in patients with hematologic malignancies. IFIs caused by filamentous fungi, especially invasive aspergillosis (IA), have become high prevalence in some groups of patients with predisposing factors, such as allogeneic stem-cell transplant (SCT) recipients [Fukuda et al. 2003; Lass-Flörl, 2009; Marr et al. 2002; Upton et al. 2007]. Drug-based strategies for IFIs include various approaches such as prophylactic, empiric, preemptive, and directed treatment. Despite its demonstrated efficacy in other infections such as pneumocystis, the use of antifungal prophylaxis in patients at high risk for IFIs remains controversial.
Among the IFIs, Aspergillus infection is associated with a particularly high rate of morbidity and mortality in immunocompromised patients [Bhatti et al. 2006]. Aspergillus conidia is found in suspension in the air and is easily inhaled. Prolonged neutropenia, prolonged use of steroids and allogeneic SCT have been classically associated with a high-risk of developing IA [Ben-Ami et al. 2009; Pagano et al. 2006]. Until recently, treatment of IA with the available antifungal agents resulted in an unacceptably high rate of treatment failure [Girmenia et al. 2012]. Most amphotericin B formulations are associated with renal toxicity precluding prolonged administration. Several new antifungal agents including new generation azoles, liposomal amphotericin B and echinocandins have recently been developed for the treatment of IA with acceptable toxicity. In addition, these new compounds obtain higher response rates.
Diagnosis of IFI is often difficult and frequently made when the disease has progressed significantly, which limits the efficacy of any antifungal therapy. Many attempts have been made to reduce the incidence of advanced IFIs using empiric or preemptive therapy without universal consensus [Cornely, 2008; Goodman et al. 1992; Marr, 2008]. Therefore, primary antifungal prophylaxis is widely used in many centers for high-risk patients. This review discusses the role of the new generation azoles, and in particular, voriconazole in the prophylaxis of IFIs in patients with hematologic malignancies.
Epidemiology of invasive fungal infections
The prophylactic use of fluconazole has been shown to reduce the incidence of invasive candidiasis (IC) and improve survival, despite its lack of activity against filamentous fungi [Cornely, 2008]. This has increased the frequency of IA among patients with hematologic malignancies in recent years. In addition, there are more patients at risk of IFI given the improvements in supportive care and the treatment of hematologic malignancies [Lass-Flörl, 2009; Pfaller and Diekema, 2007].
The incidence of IC has remained relatively stable in the USA, with an incidence rate of 22–29 per 100,000 population [Lass-Flörl, 2009; Viscoli et al. 1999]. In Europe, the incidence is lower (1.4–11 per 100,000 population) but associated with high mortality rates ranging from 24% to 59% [Kao et al. 1999; Marr et al. 2000b; Tortorano et al. 2006]. C. albicans has been typically responsible for the majority of the cases of IC, although there is a clear increase in non-albicans species, such as C. glabrata, C. parapsilosis, C. tropicalis, and C. krusei in certain geographic regions, including an increase in the azole-resistant species C. krusei and C. glabrata among bone marrow transplant recipients [Lass-Flörl, 2009]. C. krusei is innately resistant to fluconazole whereas C. glabrata can acquire azole resistance [Pagano et al. 2006].
An increasing number of IFIs are due to Aspergillus and other molds [Neofytos et al. 2009]. Pulmonary IA is the most common cause of pneumonia-related mortality in allogeneic SCT recipients and patients with leukemia [Kontoyiannis et al. 2010]. A prospective US study of 234 SCT recipients reported that IA was the most frequent IFI (59.2%), followed by IC (24.8%) [Girmenia et al. 2000; Pagano et al. 2006].
In a recent survey, zygomycosis was responsible for 8% of all mold infections in the transplant population [Barberán et al. 2011; Cornely et al. 2011; Ruiz-Camps et al. 2011]. Zygomycosis is currently considered the second most common invasive mold infection after aspergillosis and its incidence appears to be increasing among SCT recipients [Böhme et al. 1995; Ruping et al. 2008b]. Fusarium species are rare pathogens that can also cause IFIs in patients with hematologic diseases with a high associated mortality rate of 53% [Del Bono et al. 2008; Fukuda et al. 2003]. The incidence of scedosporium infection is also increasing among patients with hematologic disease with mortalities around 67% (including 100% for Scedosporium apiospermum and Scedosporium prolificans infections in some reports) [Marr et al. 2002].
Risk groups
Identifying patients at high risk for IFI is critical when designing any antifungal prophylaxis. Several proposals for risk stratification have been published [Bochud et al. 2008]. One of them stratifies patients according to several risk factors into three groups (high, intermediate and low risk) and subsequently uses risk modifiers related to the host, treatment, or environment to fine tune the classification [Altes et al. 2004; Mezger et al. 2010].
A major risk factor for IFI is prolonged and profound neutropenia [Vallejo and Len, 2010]. Specific risk factors for IC in patients with hematologic malignancies include Candida colonization, complicated abdominal surgery, parenteral nutrition, prolonged antibiotic use, and central vascular catheters [Cornely et al. 2011]. Specific risk factors for IA are hematologic malignancies such as acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) undergoing intensive induction chemotherapy [Menzin et al. 2009]. A special case can be made for patients with ALL receiving vinca alkaloids since azole antifungals are not routinely used for fear of worsening vincristine-related toxicity [Marr et al. 2000a].
IFI risk is increased following allogeneic hematopoietic SCT (HSCT), after using alternative donors or in the presence of graft versus host disease (GVHD), requiring high-dose steroids, monoclonal antibodies, or antitumor necrosis factor drugs [Marr et al. 2004a]. Certain viral infections (cytomegalovirus and respiratory viruses) are also associated with an increased risk of IFIs [Glasmacher et al. 2006]. Recently published reports suggest that donor anti-Aspergillus immunity could influence the risk of IFIs following allogeneic SCT [van Burik et al. 2004]. Host-related factors, such as iron overload or genetic polymorphisms, have also been linked to an increased risk of IFIs [Cornely et al. 2007; Ullmann et al. 2007]. These risk modifiers could change the initial risk group of a given patient [Ullmann et al. 2007]. Table 1 summarizes common risk factors for both IC and IA, Table 2 shows the risk-modifying factors, and Figure 1 shows the proposed algorithm for risk adjustment.
Overall risk factors for developing invasive fungal infection.
ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CMV, cytomegalovirus; GVHD, graft versus host disease; HEPA, high-efficiency particulate air; HSCT, hematopoietic stem-cell transplantation; IA, invasive aspergillosis; wk, week; wks, weeks.
Risk-modifying factors for developing invasive fungal infection.
Adapted from Vallejo and Len [2010].
ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; ATG, antithymocyte globulin; CMV, cytomegalovirus; GVHD, graft versus host disease; HEPA, high-efficiency particulate air; HSCT, hematopoietic stem-cell transplantation; IFI, invasive fungal infection; MDS, myelodysplastic syndrome; RSV, respiratory syncytial virus; SAA, serum amyloid A.
Risk groups for invasive fungal infections. Patients are stratified according to risk factors into high-, intermediate-, and low-risk groups. Risk modulators are used to upgrade or downgrade the risk group of individual patients. ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CBT, cord blood transplant; GVHD, graft versus host disease; HLA, human leukocyte antigen; MDS, myelodysplastic syndrome; SCT, stem-cell transplantation. (Adapted from Barberán et al. [2011].)
Antifungal strategies
There are several treatment strategies that can be used to manage IFIs: prophylactic, empiric, preemptive, and directed treatments [Cornely et al. 2007]. Prophylaxis applies to the use of antifungal drugs in all patients at risk; empiric therapy is given to high-risk patients with neutropenic fever unresponsive to broad spectrum antibacterials; preemptive therapy is the administration of antifungal agents to patients with only radiographic or laboratory based (Galactomann test, polymerase chain reaction) evidence of IFI with or without signs or symptoms of infection; and directed therapy is treatment with an appropriate antifungal when there is definitive proof of a fungal infection. Embedded into this spectrum is the intention to treat the IFI as soon as necessary without offering unnecessary treatment to patients who do not have an IFI.
Prophylaxis is an attractive strategy in high-risk patients, given the lack of reliable diagnostics and the high mortality rate associated with IFIs. Even if fungal death can be prevented with modern treatments, avoiding detrimental repercussions (i.e. delay in chemotherapy) on the treatment of the underlying disease by preventing the occurrence of IFIs may result in a survival advantage. A large retrospective database study of 11,881 patients with IFIs and matched controls showed that those with IFIs had higher mortality, a longer hospital stay, and higher hospital costs compared with otherwise similar patients who did not develop IFIs (Voriconazole-SMPC). Despite this, clear benefit from antifungal prophy laxis must be documented in clinical studies before routine adoption of antifungal prophylaxis worldwide.
The value of antifungal prophylaxis
Two large, double-blind, randomized, placebo-controlled studies conducted in the 1990s showed that fluconazole prophylaxis reduced the incidence of IFI in patients undergoing SCT [Hicheri et al. 2012]. An 8-year follow-up update confirmed that survival remained significantly improved for patients who had received fluconazole prophylaxis. In one of these studies the incidence of severe gut GVHD was also decreased [Cecil and Wenzel, 2009].
Itraconazole in prophylaxis has been compared with fluconazole in a randomized study of 304 allogeneic SCT recipients [Cecil and Wenzel, 2009; Herbrecht et al. 2002; Walsh et al. 2008]. This study demonstrated that itraconazole was superior to fluconazole against mold infections but similar against Candida with no difference in overall survival. As expected, more patients in the itraconazole group discontinued therapy because of intolerance. However, in another open-label, randomized study in 494 patients with hematologic cancers there were no differences in the efficacy or safety of itraconazole compared with fluconazole prophylaxis [Maertens et al. 2011; Walsh et al. 2008]. The superior efficacy of micafungin versus fluconazole has been shown in a randomized, double-blind, comparative phase III study in 882 SCT recipients [Wingard et al. 2010].
In two large studies posaconazole was effective against IFI [Wingard et al. 2010]. In one randomized, double-blind study comparing posaconazole with fluconazole in 600 SCT recipients with mostly chronic GVHD, posaconazole was as effective as fluconazole in preventing IFIs [Wingard et al. 2007]. In this study, patients on posaconazole prophylaxis had fewer breakthrough IFIs (including IA), although the overall mortality was not different. The incidence of treatment-related adverse events was similar in both groups.
The second randomized study of posaconazole versus fluconazole (some patients received itraconazole) included 602 patients with AML or myelodysplastic syndrome (MDS) treated with chemotherapy [Marks et al. 2011]. The results showed that fewer patients in the posaconazole group developed IA, proven or probable IFIs, although in contrast with the previous study, there was a statistically significant difference in overall survival for posaconazole-treated patients.
The rationale for voriconale as antifungal prophylaxis
An ideal antifungal for prophylaxis should have a potent and broad activity, be available both orally and intravenously, and have a low-toxicity profile. Voriconazole fulfills all these criteria.
Voriconazole is authorized for the treatment of IA, Candida infections in patients without neutropenia, and serious infections caused by Scedosporium and Fusarium species [Marks et al. 2011]. In Europe it is licensed for the treatment of fluconazole-resistant serious invasive Candida infections and in the USA for esophageal candidiasis and disseminated Candida infections of the skin, abdomen, kidney, bladder wall and wounds [Marks et al. 2011]. This corresponds with the voriconazole broad spectrum of activity against yeasts and molds, including Aspergillus, Candida, Fusarium, Scedosporium and Cryptococcus species. Voriconazole, however, is not active against zygomycetes [Marks et al. 2011].
The extended spectrum of voriconazole gives it potential value as a prophylactic agent. The clinical efficacy of voriconazole against the majority of fungal pathogens [Chabrol et al. 2010; Gergis et al. 2010; Martin et al. 2010; Torres et al. 2010; Trifilio et al. 2007] makes it potentially very useful for the prevention of IFIs in patients with hematologic malignancies. However, given that, at present, voriconazole is widely recommended as first-line treatment for proven or probable IA [Cordonnier et al. 2010; Cornely et al. 2008; El-Cheikh et al. 2010], this may result in strategic and therapeutic dilemmas if used as prophylaxis and in an increased risk of selection of resistant species.
Clinical trials of voriconale as antifungal prophylaxis
There are two large randomized clinical studies that have evaluated voriconazole as primary antifungal prophylaxis following allogeneic HSCT (Table 3). The first was a randomized, double-blind trial comparing voriconazole with fluconazole [Cordonnier et al. 2004; Cornely et al. 2008]. Patients at high risk for IFIs were excluded. Prophylaxis was administered for a minimum of 100 days, which could be extended to 180 days for patients receiving immunosuppressants.
Results of randomized controlled clinical trials of voriconazole as primary prophylaxis in patients with hematologic malignancies.
In the IMPROVIT trial FFS included a composite endpoint (no discontinuation > 14 days).
FFS, failure free survival; IFI, invasive fungal infection; OS, overall survival; RIC, reduced conditioning; SIB, sibling; URD, unrelated donor.
Serum galactomannan was used in all patients and an intensive diagnostic screening process was conducted for all positive patients. Empiric therapy was allowed for a maximum of 2 weeks. The primary endpoint of the study was fungal-free survival at 6 months post transplant [Cordonnier et al. 2004; Martino et al. 2006].
A total of 295 patients were treated with fluconazole and 305 with voriconazole. Their diagnosis included acute leukemia, chronic myeloid leukemia, MDS, and non-Hodgkin lymphomas. The median age was 43 years (2–65), although 92% of patients were adults. More than half of the donors were human leukocyte antigen (HLA) identical while only 4% of donors were HLA mismatched. Both groups were well balanced with regards to demographics, underlying disease, IFI risk, or SCT characteristics. Engraftment, acute or chronic GVHD, and nonfungal infections were not significantly different between the two groups.
The analysis of the main endpoint of the study showed that fungal-free survival at 6 months was 78% with voriconazole and 75% with fluconazole. There were also no differences in overall survival at 6 and 12 months. Similarly, adverse events were equally distributed between the two groups. Success of prophylaxis at day 180 in patients with GVHD was also significantly higher with voriconazole (50% versus 30%; p = 0.03).
The cumulative incidence of proven, probable, and possible IFI was similar between the two arms: 7.3% for voriconazole and 11.2% for fluconazole at 6 months. Aspergillus was the cause of the IFIs most frequently in patients treated with fluconazole (17 cases) than in patients treated with voriconazole (nine cases, p = 0.09) with no significant differences in other fungi. According to this study, voriconazole and fluconazole seem to have similar efficacy in decreasing mortality in recipients of allogeneic SCT. However, the authors noted a high use of empiric therapy in both groups that could have confounded the interpretation of the study [Masamoto et al. 2011].
The IMPROVIT study was a clinical trial evaluating voriconazole and itraconazole as primary prophylaxis in patients receiving allogeneic SCT in a prospective, phase III, randomized, open-label, multicentre clinical fashion [Cordonnier et al. 2010]. As before, prophylaxis was administered for a minimum of 100 days, which could be extended to 180 days for patients receiving immunosuppressants. Empiric therapy was allowed for a maximum of 2 weeks in case of possible IFIs. Patients were stratified by donor type and conditioning regimen intensity. Following intravenous loading and a minimum of 1 week intravenous administration, both voriconazole and itraconazole could be administered orally at 200 mg twice daily [Cordonnier et al. 2010].
The primary endpoint was a composite endpoint consisting of fungal-free survival to day 180 without having to discontinue the prophylactic drug longer than 14 days in total before day 100. A total of 224 patients were enrolled in the voriconazole arm and 241 in the itraconazole arm. Both groups were similar with regards to baseline characteristics (including underlying disease, conditioning regimen and T-cell depletion) and specific IFI risk factors including GVHD.
Voriconazole was superior to itraconazole for the primary endpoint (success of antifungal prophylaxis at day 180). The success rates were 49% for voriconazole and 33% for itraconazole patients (p < 0.01). Voriconazole was also superior to itraconazole at day 100. The superiority of voriconazole was mainly due to its inferior discontinuation rate with no differences in overall survival and breakthrough fungal infections [Ruping et al. 2008a; Segal et al. 2007].
The most frequent reasons for treatment discontinuation were adverse events (23.2%) and study drug intolerance (21.6%) in the case of itraconazole and the occurrence of adverse events (29.9%) with voriconazole. Itraconazole-related side effects were typically gastrointestinal intolerance, while voriconazole produced more liver enzyme abnormalities. Itraconazole-treated patients had almost 30 days less of prophylaxis compared with voriconazole, with a median treatment duration of 68 days and 96 days respectively (p < 0.01). These results suggest that voriconazole has better long-term treatment tolerance than itraconazole, which is likely to result in a major advantage in patients (such as allogeneic SCT recipients or patients with GVHD) who require long-term prophylaxis.
Both groups had identical overall survival at day 100 (92% in each group) and day 180 (82% for voriconazole and 81% for itraconazole). There were no statistically significant differences with regards to proven or probable IFIs or breakthrough IFIs. This study showed a remarkably low rate of IFIs, with only eight IFIs diagnosed during the trial. In the voriconazole group there were three IFIs (one aspergillosis and two candidemia) compared with five IFIs in the itraconazole group (all aspergillosis). There were no cases of invasive zygomycosis in either group.
Empiric antifungal therapies, including caspofungin and liposomal amphotericin B, were used more frequently in the itraconazole arm (42% versus 30%; p < 0.01). Significantly up to 14% of patients in the itraconazole arm received voriconazole as empiric therapy during the study. These data make the interpretation of the results more difficult and may even underscore the advantage of voriconazole over itraconazole. The authors concluded that primary prophylaxis with voriconazole effectively prevents IFIs following allogeneic HSCT, with acceptable safety. The superiority of voriconazole in the primary composite endpoint of the study appeared to be due mainly to better long-term tolerability [Cordonnier et al. 2010].
A number of other recent studies suggest that voriconazole is efficacious as antifungal prophylaxis in HSCT recipients and patients with acute leukemia [Vehreschild et al. 2008].
Role of voriconazole in secondary prophylaxis
In patients with hematologic malignancies and a previous IFI there is a high risk of recurrence or the development of a new IFI following new chemotherapy or SCT [Lamaris et al. 2008]. The probability of relapse of the IFI following allogeneic HSCT is not well known but varies between 19% and 33% and between 16% and 52% in patients with AML who undergo intensive chemotherapy [Chandrasekar, 2009; Han et al. 2010; Marr, 2008; Pascual et al. 2008]. The relapse risk is even higher in cases of partial resolution of the previous IFI, new prolonged neutropenia, and use of high-dose cytarabine [Ruping et al. 2011]. There is some evidence that administration of broad-spectrum antifungal agents could be beneficial in the prevention of IFI relapse or progression, therefore allowing for the treatment to continue in the presence of residual disease [Pascual et al. 2008, 2012].
A prospective, open-label, multicentre study of secondary prophylaxis with voriconazole has been published recently [Bruggemann et al. 2009; Cecil and Wenzel, 2009; Cronin and Chandrasekar, 2010; Pascual et al. 2012]. It enrolled 45 patients with a prior IFI (proven or probable) less than 12 months before entering the trial. All patients underwent allogeneic HSCT for a variety of hematologic conditions. Exclusion criteria included active IFI, a history of zygomycosis, previous failure of voriconazole in antifungal therapy, and liver or renal insufficiency. Voriconazole was commenced at least 2 days after the end of the conditioning regimen and was planned for 100 days after transplant. The primary efficacy variable of the study was the incidence of proven/probable IFI during 12 months.
Almost all patients (41/45) had a diagnosis of acute leukemia (53% in first complete remission). Previous IFIs were proven IA (n = 6), probable IA (n = 25), proven IC (n = 5), other proven IFI (n = 3), or other probable IFI (n = 3). In three patients, a previous systemic fungal infection could not be confirmed. A total of 35 patients had received voriconazole as treatment for the previous IFI. The median time between resolution of previous IFI and the date of transplant was 59 days (range 3–311). The median duration of voriconazole prophylaxis was 94 days (range 5–180). Although this was a high-risk population [there were 11 deaths (24%) in the12 months after transplant], only one patient died from a IFI. A total of three episodes of IFI were reported: two relapses of the previous IFI (one Candida and one Scedosporium) and one new episode (zygomycosis in a patient with previous IA). Significantly, no case of IA was observed after transplant in this study. The cumulative incidence of IFIs was 7% at 12 months, considerably lower than expected [Marr et al. 2004b].
Integrating voriconazole prophylaxis into an antifungal strategy
As discussed before, there are various treatment strategies that can be used to manage IFI: prophylactic, empiric, preemptive, and directed. The choice of prophylaxis is likely to influence the drugs and strategies selected thereafter [Cecil and Wenzel, 2009; Cronin and Chandrasekar, 2010; Pascual et al. 2012; Rogers and Frost, 2009]. For instance, fluconazole prophylaxis does not include molds so there is a need for a rapid reaction to cover IFI by molds in the empiric therapy stage.
However, voriconazole is highly effective against molds, and as reported above, is capable of reducing the incidence of IFIs to less than 10% in patients with hematologic diseases (less than 3% in the IMPROVIT trial) [Feist et al. 2012; Singer et al. 2012; Vadnerkar et al. 2010; Vanacker et al. 2008]. Therefore, some centers may decide not to use empiric treatment and opt for a purely preemptive strategy [Fera et al. 2009; Garcia-Effron et al. 2010]. The choice of antifungal agent could also be influenced by the use of voriconazole in prophylaxis: azole-resistant fungi should be taken into consideration when choosing the appropriate drug.
Finally, once an IFI has been proven, there is uncertainty about the continuous use of voriconazole for azole-sensitive microorganisms (such as Aspergillus) in patients who have received prophylaxis with the same agent [Fera et al. 2009].
Monitoring voriconazole plasma levels
Monitoring plasma levels is advised when individualization of drug dose is required to ensure safety and efficacy. Oral bioavailability of voriconazole has been claimed to be close to 100% so that no dose adjustment is necessary when switching from intravenous to oral administration [Fera et al. 2009; Snelders et al. 2008; Verweij et al. 2009]. Following administration of oral voriconazole as antifungal prophylaxis, the agent seems to exhibit good tissue penetration into the lungs, the most common primary infection site in IA [Pfaller et al. 2012]. Recently, the dose of voriconazole has been challenged by several investigators to recommend a slightly higher dose of 300 mg twice daily to achieve therapeutic plasma levels in the majority of patients [Pfaller and Diekema, 2007]. However, routine availability of voriconazole plasma levels is not widespread.
However, in most patients with hematologic diseases there is the potential for pharmacokinetic drug–drug interactions given that voriconazole is metabolized through the cytochrome P450 system [Pongas et al. 2009]. For example, interactions between liver-metabolized chemotherapy agents (e.g. vinca alkaloids and anthracyclines) and voriconazole may result in unacceptable toxicity, unless appropriate dose adjustments can be established [Verweij et al. 2009]. Studies have also shown that in combination with cyclophosphamide, itraconazole may cause clinically significant interactions, including hypertension, neurotoxicity, and gastrointestinal toxicity [Snelders et al. 2008]. However, similar interactions have so far not been reported for voriconazole.
Voriconazole has several potential benefits as a prophylactic agent; however, its use is also associated with a number of concerns. There is the possibility of serious adverse events, such as prolonged visual disturbances, QT-interval prolongation and hepatic toxicity, therefore close monitoring of cardiac, visual, and liver function is strongly recommended [Ananda-Rajah et al. 2008; Chang et al. 2008; Cornely et al. 2008; Vande Broek and Schots, 2009]. In addition, there have been a small number of reports of squamous cell carcinoma as well as melanoma during long-term voriconazole treatment [Bruggemann et al. 2009; Pascual et al. 2008; Rogers et al. 2011], potentially associated with the photosensitivity effect of this agent. However, the contribution of voriconazole to the development of squamous cell carcinoma has not been fully established.
Development of resistance to voriconazole
Every antifungal drug is theoretically exposed to the development of resistance by fungal organisms. Caspofungin resistance of Candida species is linked to mutations in the Candida FKS1 and FKS2 genes [Trifilio et al. 2007]. Triazole resistance can result from altered cellular accumulation of the azole, increased levels or decreased affinity to the azole cellular target (Erg11p), and modification of the ergosterol biosynthesis pathway [Girmenia et al. 2012]. Aspergillus azole resistance may be specifically caused by a substitution of leucine 98 for histidine in the cyp51A gene, together with two tandem copies of a 34 bp sequence in the gene promoter (TR/L98H) [Bohme et al. 2009; Maertens et al. 2011; Maschmeyer et al. 2009; Walsh et al. 2008].
The SENTRY Antifungal Surveillance Program analysis of a large number of blood isolates of Candida species in the USA and Europe showed that resistance to fluconazole and itraconazole were common among C. glabrata and C. krusei isolates. The ARTEMIS DISK Surveillance Program (1997–2003) showed in 2003 that fluconazole resistance could be as frequent as 80% for C. krusei.
In recent years concerns about the advent of voriconazole-resistant Aspergillus species have proved to be unsubstantiated. In addition, azole resistance as a consequence of the widespread use of azole fungicides (such as propiconazole in crop fumigation) in the environment is very uncommon. Unlike bacteria, fungi do not transfer drug resistance genetically to other fungi, and most pathogenic fungi do not reproduce sexually in humans.
The management of breakthrough fungal infections
Patients receiving antifungal prophylaxis with voriconazole who develop clinical signs or symptoms of infection should be evaluated for breakthrough IFI. Measurement of azole levels should be considered in patients who have documented breakthrough IFI. In some studies breakthrough IFIs have been shown to occur in patients with a voriconazole trough plasma level up to 2 μg/ml.
There are no clinical studies addressing the optimal management of patients with breakthrough IFIs while on voriconazole prophylaxis. Few alternative options with similar efficacy and tolerability to voriconazole exist for the treatment of breakthrough IA. In general, it is recommended to switch to another class of mold-active antifungals, preferably liposomal amphotericin B. Although few data are available, combination therapy with different classes of antifungals may be able to improve outcome and prognosis in high-risk patients with refractory infections. A prospective, randomized study comparing the efficacy of an anidulafungin/voriconazole combination with that of voriconazole alone for IA has recently been completed, although so far it has only been presented in abstract form.
Conclusion
At present, voriconazole appears to be very effective for the primary and secondary prevention of IFIs in patients with hematologic diseases and those receiving allogeneic HSCT. Voriconazole is generally safe and well tolerated. Randomized controlled trials evaluating voriconazole as primary antifungal prophylaxis in patients with neutropenia treated for a variety of hematologic malignancies have been performed, confirming its value as a prophylactic agent. Furthermore, prospective studies assessing the possible impact of voriconazole prophylaxis on the incidence of resistant breakthrough fungal infections such as zygomycosis are clearly needed, as are studies to establish the potential prophylactic/therapeutic range of voriconazole plasma levels and possible dose adjustments during concomitant treatment with chemotherapy or other potential drug interactions.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
The authors declare no conflicts of interest in preparing this article.