ABSTRACT
The in vitro activity of the novel triazole antifungal agent posaconazole (Noxafil; SCH 56592) was assessed in 45 laboratories against approximately 19,000 clinically important strains of yeasts and molds. The activity of posaconazole was compared with those of itraconazole, fluconazole, voriconazole, and amphotericin B against subsets of the isolates. Strains were tested utilizing Clinical and Laboratory Standards Institute broth microdilution methods using RPMI 1640 medium (except for amphotericin B, which was frequently tested in antibiotic medium 3). MICs were determined at the recommended endpoints and time intervals. Against all fungi in the database (22,850 MICs), the MIC50 and MIC90 values for posaconazole were 0.063 μg/ml and 1 μg/ml, respectively. MIC90 values against all yeasts (18,351 MICs) and molds (4,499 MICs) were both 1 μg/ml. In comparative studies against subsets of the isolates, posaconazole was more active than, or within 1 dilution of, the comparator drugs itraconazole, fluconazole, voriconazole, and amphotericin B against approximately 7,000 isolates of Candida and Cryptococcus spp. Against all molds (1,702 MICs, including 1,423 MICs for Aspergillus isolates), posaconazole was more active than or equal to the comparator drugs in almost every category. Posaconazole was active against isolates of Candida and Aspergillus spp. that exhibit resistance to fluconazole, voriconazole, and amphotericin B and was much more active than the other triazoles against zygomycetes. Posaconazole exhibited potent antifungal activity against a wide variety of clinically important fungal pathogens and was frequently more active than other azoles and amphotericin B.
Over the past 2 decades, the incidence of systemic mycoses has increased dramatically. This is primarily due to the increase in the number of at-risk individuals, principally those with impaired immunity, such as transplant recipients, cancer patients receiving chemotherapy, and human immunodeficiency virus-infected patients (2, 17, 24, 32, 37). The most common fungal pathogens are species of Candida, Cryptococcus, Coccidioides, Aspergillus, and Histoplasma; less common pathogens include agents of zygomycosis (primarily species of Rhizopus, Mucor, Cunninghamella, Apophysomyces, Absidia, and Rhizomucor), hyalohyphomycosis, and phaeohyphomycosis (32).
Mortality rates associated with systemic mycoses, particularly those involving members of the zygomycetes, remain unacceptably high. Effective treatment requires both an early diagnosis, to facilitate prompt initiation of therapy, and broad-spectrum therapeutic agents with activity against both common and “emerging” pathogens. Until recently, the drugs available to treat invasive fungal infections were limited by their spectrum of activity, the development of resistance, and less than optimal tolerability and drug interaction profiles (15). To address these issues, a new generation of triazoles, including posaconazole (POS), voriconazole (VRC), and ravuconazole (RAV), has been developed. These agents possess potent broad-spectrum activity and favorable pharmacokinetic profiles (3, 12, 15). Among these extended-spectrum triazoles, POS has proven to be a potent inhibitor of ergosterol synthesis in both yeasts and molds (19) and to be active against a wide range of pathogens (1, 4, 28, 29), including Aspergillus spp. (16, 29) and the zygomycetes (7, 34).
This report summarizes in vitro data for 19,000 clinically important strains of yeasts and molds collected from 200 medical centers worldwide over a 10-year time span. Where available, data are also provided on the comparator drugs itraconazole (ITC), fluconazole (FLC), VRC, and amphotericin B (AMB). Overall, POS exhibited potent broad-spectrum antifungal activity; it was frequently more active than the other azoles, and its spectrum of activity was comparable to that of AMB and superior to those of all other marketed antifungals.
(Part of this work was presented at the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., 30 October to 2 November 2004.)
MATERIALS AND METHODS
Antifungal agents.POS was prepared at Schering-Plough Research Institute, Kenilworth, NJ. ITC and AMB were obtained from Janssen Pharmaceutica N.V., Beerse, Belgium, and Sigma Chemical Co., St. Louis, MO, respectively. VRC and FLC were obtained from Pfizer Inc., New York, NY.
Susceptibility testing.MIC testing was performed as described in the Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS) documents M27-A2 and M38-A and versions thereof (20, 21). For slower-growing organisms, such as the dermatophytes, Cryptococcus and Histoplasma spp., if insufficient growth was observed at 48 h then the plates were incubated for longer periods (typically 72 h). Test panels were either prepared in the individual laboratories using drug powders or obtained as frozen panels from Trek Diagnostics Systems Inc. (Cleveland, OH).
Data analysis.Susceptibility data were collected from individual investigators and entered into a global database. Not all strains were tested against all of the comparator drugs; however, all strains in this study were tested against POS. All data relating to control/quality control isolates were excluded from the analysis. In a few instances, an investigator may have tested an isolate more than once; consequently, in the tables, “n” refers to the number of MICs not the number of isolates.
RESULTS
All isolates.Overall, POS exhibited potent in vitro activity against approximately 19,000 fungal microorganisms. MIC50 and MIC90 values for POS were as follows: 0.063 μg/ml and 1.0 μg/ml, respectively, for all fungi (22,850 MICs); 0.125 μg/ml and 1.0 μg/ml for all molds (4,499 MICs); and 0.063 μg/ml and 1.0 μg/ml for all yeasts (18,351 MICs).
For the subsets of these 19,000 isolates that were also tested against other antifungal agents, POS was more active than, or within 1 dilution of, ITC, FLC, VRC, and AMB (Table 1). Although VRC exhibited a lower mean MIC50 than did POS against yeasts, POS was more active than VRC against molds.
In vitro activities of posaconazole, itraconazole, fluconazole, voriconazole, and amphotericin B against all fungi, molds, and yeasts tested
Mold isolates.For subsets of mold isolates tested against each antifungal agent, POS was either more potent than or equivalent to ITC, AMB, and VRC (Table 2). Against hyaline molds, including Aspergillus spp., Fusarium spp., and a miscellaneous group comprising other species, such as Acremonium, Basidiomycetes, Bjerkandera, Coprinus, Paecilomyces, Pseudallescheria, and Schizophyllum, POS was equivalent to VRC, AMB, and ITC.
Comparative in vitro activities of posaconazole, itraconazole, voriconazole, and amphotericin B against mold isolates
POS showed good activity against Aspergillus spp. (including A. fumigatus, A. flavus, and A. niger) and against the majority of zygomycetes (including Rhizopus, Mucor, Absidia, and Cunninghamella spp.). For all strains of Aspergillus spp. tested, the MIC50 and MIC90 values were 0.125 μg/ml and 0.5 μg/ml, respectively, whereas for all zygomycetes tested, the MIC50 and MIC90 values were 0.5 μg/ml and 4.0 μg/ml, respectively. In comparisons with other antifungal agents against Aspergillus spp. (1,423 MICs), POS was either more potent than or equivalent to ITC, VRC, and AMB (Table 2). However, POS was the only triazole that provided consistent activity against the zygomycetes (86 MICs) (Table 2).
Against dimorphic fungi (including Penicillium, Histoplasma, Blastomyces, and Coccidioides spp.), POS was generally more potent than, or equivalent to, ITC and AMB (Table 2). All drugs had limited activity against Fusarium spp. (Table 2). The Fusarium strains most susceptible to POS were F. moniliforme and F. oxysporum.
POS also showed good activity against agents that cause chromoblastomycosis, mycetoma, and phaeohyphomycosis, including Scedosporium apiospermum (though not Scedosporium prolificans) and Exophiala, Alternaria, and Bipolaris spp. (Table 2), and POS was generally more active than ITC and AMB against these organisms. Against dermatophytes, including Trichophyton rubrum, T. mentagrophytes, and T. tonsurans, POS was more potent than FLC and comparable to ITC (Table 3).
Comparative in vitro activities of posaconazole, itraconazole, and fluconazole against isolates of dermatophytes
Yeast isolates.POS showed good activity against Candida spp. (Table 4), including those species that are inherently less susceptible to FLC (e.g., Candida spp. C. glabrata, C. krusei, C. guilliermondii, and C. dubliniensis). The strains most susceptible to POS were C. albicans and C. dubliniensis, whereas C. glabrata was the least susceptible. Although POS was slightly less active than VRC against Candida spp., it was more active than either ITC or AMB. Against Cryptococcus spp., POS was more active than FLC and comparable to ITC, VRC, and AMB (Table 4).
Comparative in vitro activities of posaconazole, itraconazole, fluconazole, voriconazole, and amphotericin B against isolates of Candida spp. and Cryptococcus spp.
Azole-resistant Candida isolates.Candida isolates with MICs of >32 μg/ml, >0.5 μg/ml, and >2 μg/ml for FLC, ITC, and VRC, respectively, are considered resistant (21). Of the 6,595 isolates tested against all four azoles, 6.4%, 16.5%, and 3.3% were resistant to FLC, ITC, and VRC, respectively (Table 5). The frequency of isolates with MICs for POS that were >2 μg/ml was 3%. Resistance to one azole significantly impacted susceptibility to the other azoles.
Comparative in vitro activities of posaconazole, itraconazole, fluconazole, and voriconazole, against isolates of Candida spp. exhibiting resistance to itraconazole, fluconazole, and voriconazole
DISCUSSION
The present study has extended the findings of earlier in vitro investigations of the antifungal activity of POS in demonstrating its wide spectrum of activity against more than 19,000 strains of yeasts and molds encountered in infectious disease practice at more than 200 medical centers throughout the world. As well as having good activity against most Candida spp. (including C. glabrata and C. krusei), POS exhibited good activity against the majority of organisms responsible for causing aspergillosis, cryptococcosis, zygomycosis, chromoblastomycosis, mycetoma, and phaeohyphomycosis. In comparison with the other antifungal agents tested (FLC, ITC, VRC, and AMB), POS was generally more potent than FLC and either equipotent to or more potent than ITC, VRC, and AMB. Although POS exhibited slightly higher mean MIC50 values compared with VRC against Candida spp., including the inherently less susceptible strains C. glabrata and C. krusei, and against Cryptococcus spp., it was generally more active than VRC against molds. Against the zygomycetes, POS was the only triazole that exhibited consistent activity, but it was generally less active against these organisms than AMB. All drugs had limited activity against Fusarium spp. However, successful outcomes have been reported in patients with fusariosis who were treated with POS, suggesting that in vitro testing might not accurately predict the clinical outcome (11, 31).
Previous studies comparing the in vitro activity of POS with that of other antifungal agents have described similar findings. In comparison with other triazole agents, POS has generally been reported to have greater activity than FLC and ITC against yeasts such as Candida spp., Cryptococcus spp., and Saccharomyces cerevisiae (1, 4, 13, 23, 27, 28, 30), although in some studies, it was no more active than ITC against Candida spp. (26) or Cryptococcus neoformans (25). POS has also proved more active than AMB and flucytosine against most Candida spp. (26) and has been found to have similar activity to VRC against the majority of Candida spp. (23, 30). However, against C. glabrata, which has proved the least-susceptible Candida species to POS (28, 30), it was slightly less active than VRC, both in the present study and in an earlier investigation by Pfaller et al. (30).
Consistent with previous reports (22, 28), isolates with elevated MICs to one azole were generally less susceptible to all azoles. C. albicans and C. glabrata, in approximately equal numbers, were the species most frequently characterized as being resistant to FLC and VRC. In contrast, the majority of ITC-resistant isolates were C. glabrata. Comparing POS and VRC, the numbers of C. glabrata MICs that were >2 μg/ml (the VRC-resistant breakpoint) were nearly identical for both drugs. However, for both C. albicans and other species of Candida, the number of POS MICs that were >2 μg/ml was nearly twofold lower than for VRC.
In studies focusing on Aspergillus spp., POS has proved more active than both ITC (4, 22) and AMB (22). In a comparison of POS with RAV, VRC, ITC, and AMB against 239 isolates of Aspergillus spp. and other filamentous fungi (including Fusarium, Rhizopus, and Mucor spp.), POS was the most active agent (94% of isolates inhibited at a MIC of ≤1 μg/ml) (29). In the case of zygomycetes, POS exhibited good activity against 36 zygomycetes belonging to six genera; AMB also showed good activity, VRC was significantly less active, and ITC and terbinafine showed variable activity (7). Two additional studies compared the activity of POS with those of AMB, VRC, FLC, and ITC or with VRC and caspofungin (CSP) against collections of 37 and 59 zygomycetes, respectively (8, 34). In both studies, POS was significantly more active than VRC; in the individual studies, POS was far more active than either FLC (34) or CSP (8) and slightly more active than ITC (34). In the clinic, POS has been used as salvage therapy to treat over 100 patients with zygomycosis; the rate of success (i.e., either complete or partial response) was at least 60% (10, 35).
In agreement with our findings, good activity against Coccidioides immitis has been reported in other studies, although POS proved slightly less active than ITC in one study (9). Both this and previous studies demonstrated that POS is less active against Scedosporium prolificans than against S. apiospermum (5). Similarly, although POS was not compared with VRC against these organisms in the data presented above, other investigators have shown that POS is significantly less active than VRC against S. prolificans and slightly less active than VRC against S. apiospermum (5, 18).
The molecular basis for the enhanced in vitro activity of POS over the other azoles remains to be determined. At a first approximation, the in vitro activity of a drug is governed by its ability to accumulate within the cell coupled with its affinity for its target site. Several lines of evidence suggest that decreased susceptibility to azoles results from both changes in intracellular accumulation and changes in the target site (6, 14, 33). The azole target site is 14α-demethylase (CYP51), which is located predominantly in the endoplasmic reticulum. None of the fungal CYP51 enzymes have been crystallized; therefore, information on the way in which the azoles bind to the protein has come primarily from homology modeling studies. One recent study suggested that the long side chain of POS and ITC, a side chain that is absent in VRC and FLC, helps stabilize binding of these azoles to CYP51; this appears to be particularly true for CYP51 proteins with mutations close to the active site (36). This model also suggested that mutations that interfered with binding of the long side chain negatively impacted POS and ITC more than they impacted FLC and VRC. It is conceivable that an increased affinity for CYP51 is responsible for the unique activity of POS against the zygomycetes. In this regard, expression of the CYP51 from Rhizopus oryzae in an azole-susceptible Saccharomyces cerevisiae strain resulted in a 4-fold decrease in susceptibility to POS and a >250-fold decrease in susceptibility to VRC; there were no changes in susceptibility to either AMB or CSP (unpublished data). These data suggest that for R. oryzae, and possibly for other zygomycetes, the nature of the interaction between drug and target protein is a major determinant of susceptibility. With regard to drug accumulation, the level of efflux pump expression can strongly influence the susceptibility of a cell to azoles (33). Studies, primarily using yeasts, demonstrated that whereas all azoles appear to be substrates for the ATP-dependent pumps, POS and ITC are not substrates for the major facilitator encoded by MDR1 (D. Sanglard, F. Ischer, and J. Bille, Abstr. 42nd Intersci. Conf. Antimicrob. Agents Chemother., abstr. M-221, p. 379, 2002); again, the molecular basis for these differences remains to be established.
In summary, the differences between POS and the other triazoles described above may account for the unusually broad spectrum of activity of POS and may also be important in combating the increase in triazole resistance currently being observed among some fungal pathogens, notably Candida spp., for which multiple molecular mechanisms may be responsible for the decrease in susceptibility.
Conclusion.Overall, POS exhibited potent antifungal activity and had a broad spectrum of activity. POS was more potent than FLC against all organisms tested and was frequently more potent than ITC, VRC, and AMB. Among the triazoles, POS was the only agent that exhibited consistent activity against the zygomycetes. POS also showed good activity against the vast majority of organisms that cause aspergillosis, candidiasis, cryptococcosis, chromoblastomycosis, mycetoma, and phaeohyphomycosis, confirming its potential as a useful agent for patients with serious systemic mycoses.
ACKNOWLEDGMENTS
The authors would like to thank the following principal investigators at testing sites who contributed susceptibility data to the SPRI database. The investigators from the United States were as follows: David Andes, University of Wisconsin, Madison; John Galgiani, Valley Fever Center for Excellence, Tucson, Arizona; Mahmoud Ghannoum, University Hospitals of Cleveland and Case Western Reserve University, Cleveland, Ohio; John Graybill, Thomas Patterson, Sofia Perrea, Michael Rinaldi, and Stephen Sanche, University of Texas Health Science Center at San Antonio; Geraldine Hall, The Cleveland Clinic Foundation, Cleveland, Ohio; Duane R. Hospenthal, Brooke Army Medical Center, Fort Sam Houston, Texas; Ana Espinel-Ingroff, Medical College of Virginia, Virginia Commonwealth University, Richmond; Elias Manavathu, Wayne State University, Detroit, Michigan; John Perfect, Duke University Medical Center, Durham, North Carolina; Michael Pfaller and Daniel Diekema, University of Iowa College of Medicine, Iowa City; John Rex and Luis Ostrosky-Zeichner, University of Texas Medical School at Houston; Glenn Roberts and Arthur Guruswamy, Mayo Clinic, Rochester, Minnesota; Alan Sugar, Boston University Medical Center, Boston, Massachusetts; Thomas Walsh and Ruta Petraitiene, National Cancer Institute, National Institutes of Health, Bethesda, Maryland; and Joseph Wheat, MiraVista Diagnostics, Indianapolis, Indiana. The investigators from the rest of the world were as follows: Hail Al-Abdely, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia; Mickael Aoun, Institut Jules Bordet, Brussels, Belgium; Sevtap Arikan, Hacettepe University Medical School, Ankara, Turkey; Francesco Barchiesi, University of Ancona, Ancona, Italy; Luis Carrasco, University of Madrid, Madrid, Spain; Eric Dannaoui, Université Claude Bernard Lyon I, Lyon, France; Bertrand Dupont, Institut Pasteur, Paris, France; Miguel Gobernado, Hospital Universitario “La Fe” Valencia, Spain; Gloria González, Universidad Autónoma de Nuevo León, Nuevo León, Mexico; Aditya Gupta, University of Toronto, Toronto, Canada; Marie-Pierre Hayette, Université de Liège, Liège, Belgium; Michel Laverdière, Hôpital Maisonneuve-Rosemont, Montreal, Quebec, Canada; Frank-Michael C. Müller, University of Heidelberg, Heidelberg, Germany; Alfonso Carrillo-Muñoz, ACIA Microbiología, Barcelona, Spain; Marcio Nucci, University Hospital, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil; Juan L. Rodriguez-Tudela and Manuel Cuenca-Estrella, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Madrid, Spain; M. Carmen Rubio, Lozano Blesa University Hospital, Zaragoza, Spain; Anna Maria Tortorano, Università degli Studi—IRCCS Ospedale Maggiore, Milan, Italy; Katsuhisa Uchida, Teikyo University Institute of Medical Mycology, Tokyo, Japan; Paul Verweij, University Medical Center St. Radboud, Nijmegen, The Netherlands; Miriam Weinberger, Rabin Medical Center, Tel Aviv University, Petach Tikva, Israel; S. T. Yildiran, Gulhane Military Medical Academy and School of Medicine, Turkey; and AB Biodisk, Solna, Sweden.
FOOTNOTES
- Received 7 February 2006.
- Returned for modification 1 March 2006.
- Accepted 1 April 2006.
- American Society for Microbiology
