INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Nystatin is a polyene antibiotic derived from Streptomyces noursei (12). It is active against a broad spectrum of fungi in vitro and in vivo, including Aspergillus fumigatus, Candida albicans, Coccidioides immitis, Cryptococcus neoformans, and Histoplasma capsulatum (4, 5, 7, 10, 26, 27). However, nystatin is not well absorbed from the gastrointestinal tract, and its parenteral administration results in dose-limiting toxicities and harmful infusion-related reactions (21, 23). For this reason, the clinical application of nystatin has largely been limited to topical use in mucosal and cutaneous forms of candidiasis (25).

Like nystatin, amphotericin B is a polyene antibiotic with a broad spectrum of antifungal action. It remains the most effective agent currently available for the treatment of many systemic fungal infections, despite the fact that its clinical use is seriously limited by nephrotoxicity and other side effects (8). To overcome these problems, three new lipid-based parenteral formulations of amphotericin B have been developed (Table 1). These industrial preparations, which differ in their compositions and physicochemical properties, are less toxic than the conventional deoxycholate formulation of amphotericin B, and this has enabled higher doses to be administered to patients with systemic fungal infections (13).

The success of the new formulations of amphotericin B has stimulated interest in the development of less toxic parenteral preparations of nystatin. Initial attempts to develop a liposomal formulation of this agent resulted in a preparation which was much less toxic in vivo (19), as well as being active against a range of fungi in vitro and against C. albicans in vivo (18, 19). Nyotran (Aronex Pharmaceuticals Inc.) is a new multilamellar liposomal formulation of nystatin that contains dimyristoyl- phosphatidylcholine (DMPC) and dimyristoylphosphatidyl- glycerol (DMPG) in a 7:3 ratio. It has been found to be effective and well tolerated in the treatment of neutropenic animals with disseminated Aspergillus infection (9, 29). It is currently undergoing clinical evaluation in patients with systemic fungal infections and has been reported to be active in some neutropenic individuals for whom amphotericin B treatment failed (2).

In this study, we compared the in vitro activity of liposomal nystatin with those of free nystatin and four pharmaceutical preparations of amphotericin B against 200 isolates of Aspergillus spp., Candida spp., and C. neoformans. The in vitro testing method we employed was a microdilution adaptation of the standard broth macrodilution reference method of the National Committee for Clinical Laboratory Standards (20).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Test isolates. A total of 200 isolates were tested. These comprised 10 of Aspergillus flavus, 30 of A. fumigatus, 40 of C. albicans, 20 of Candida glabrata, 10 of Candida kefyr, 20 of Candida krusei, 10 of Candida lusitaniae, 20 of Candida parapsilosis, 20 of Candida tropicalis, and 20 of C. neoformans. Of the 200 isolates, 162 were recent clinical isolates submitted to the Mycology Reference Laboratory, Bristol, United Kingdom, for identification and 38 were obtained from the United Kingdom National Collection of Pathogenic Fungi, held at the Mycology Reference Laboratory. Yeast isolates were identified to the species level by the Auxacolor (Sanofi Diagnostics Pasteur, Paris, France) and API 20C (bioMerieux UK Limited, Basingstoke, England) yeast identification systems and by morphology on Oxoid cornmeal agar plates (Unipath Limited, Basingstoke, England). Two reference strains, C. parapsilosis ATCC 90018 and C. krusei ATCC 6258, were included in each batch of broth microdilution tests to ensure quality control.

Antifungal agents. Pharmaceutical preparations of five antifungal agents were obtained from the respective manufacturers: liposomal nystatin (Nyotran; Aronex Pharmaceuticals Inc.), amphotericin B deoxycholate (Fungizone; Bristol-Myers Squibb Pharmaceuticals Ltd., Hounslow, England), liposomal amphotericin B (AmBisome; Nexstar Pharmaceuticals Ltd., Cambridge, England), amphotericin B colloidal dispersion (ABCD) (Amphocil; Zeneca Ltd., Wilmslow, England), and amphotericin B lipid complex (ABLC) (Abelcet; Liposome Co. Ltd., London, England). Nystatin powder (analytical grade) was obtained from Sigma Chemical Co. (St. Louis, Mo.).

Antifungal susceptibility testing. Broth microdilution MICs were determined according to National Committee for Clinical Laboratory Standards recommendations (20). The antifungal agents were tested over a final concentration range of 0.015 to 8 µg/ml. Testing was performed in 96-well round-bottom microtiter plates. Cell suspensions of Candida spp. and C. neoformans were prepared in RPMI 1640 medium and adjusted to give a final inoculum concentration of 0.5 × 10³ to 2.5 × 10³ cells/ml. Spore suspensions of Aspergillus spp. were prepared in RPMI 1640 medium and adjusted to a final concentration of 0.4 × 10⁴ to 5 × 10⁴ spores/ml. The plates were incubated at 35°C and read after 48 h. The MIC was defined as the lowest concentration at which there was complete inhibition of growth.

Determination of MLC. The minimum lethal concentration (MLC) was determined after 48 h of incubation by removing 10 µl of the contents from all wells showing no visible growth and spreading them onto Sabouraud dextrose agar plates. The plates were incubated at 35°C for 48 h (Aspergillus and Candida spp.) or 72 h (C. neoformans). The MLC was defined as the lowest concentration at which 95% of the inoculum was killed.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Table 2 summarizes the in vitro susceptibilities of the 200 isolates of Aspergillus spp., Candida spp., and C. neoformans to the six different formulations of nystatin and amphotericin B. The data are reported as MIC and MLC ranges and the MICs and MLCs at which 50 and 90% of the isolates were inhibited or killed. In each batch of tests, the MICs for the quality control strains were within the accepted limits.

The two preparations of nystatin were more active in vitro than liposomal amphotericin B against the 10 species tested but less active than ABLC or amphotericin B deoxycholate against 9 of the species tested. In the case of A. flavus, both forms of nystatin were as active as ABLC but less active than amphotericin B deoxycholate. Liposomal nystatin was more active than free nystatin: in most cases, the MIC₅₀s, MIC₉₀s, MLC₅₀s, and MLC₉₀s were 1 to 2 doubling dilutions lower than those of free nystatin.

Of the four different amphotericin B preparations tested, ABLC appeared to be the most active in vitro, and liposomal amphotericin B appeared to be the least active in vitro. In most cases, the MIC₅₀s, MIC₉₀s, MLC₅₀s, and MLC₉₀s of ABLC were identical to, or lower than, those of amphotericin B deoxycholate for 9 of the 10 species tested (the exception being A. flavus), while those of ABCD and liposomal amphotericin B were 2 to 4 doubling dilutions higher than those of amphotericin B deoxycholate.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Nystatin has been a useful antifungal agent since the 1950s. However, toxic side effects following parenteral administration have limited its clinical application to the topical treatment of otomycosis and mucosal and cutaneous forms of candidiasis (25). The results of this investigation confirm and extend those of earlier reports which indicated that nystatin is a broad-spectrum antifungal agent which is active in vitro and in vivo against Aspergillus spp. (9, 27, 29), Candida spp. (4, 18, 19), and C. neoformans (26). Other reports have demonstrated that nystatin is effective against H. capsulatum in vitro and in vivo (5, 7) and against C. immitis in vivo (10).

Nyotran is a new multilamellar liposomal formulation of nystatin in which the phospholipid component consists of DMPC and DMPG at a molar ratio of 7:3 (Table 1). The nystatin content of this formulation is 7 mol%. The phospholipid component of Nyotran is identical to that of the earlier multilamellar formulation of amphotericin B developed by Lopez-Berestein et al. (17), which contained amphotericin B at a concentration of 5 to 10 mol%. Increasing the amphotericin B content of that formulation to 25 to 50 mol% resulted in the complete loss of defined liposomal structures and their replacement with ribbon-like structures (15). This form of lipid-based amphotericin B is now marketed as ABLC (Abelcet).

Our results indicate that liposomal nystatin is more active in vitro than free nystatin against Aspergillus spp., Candida spp., and C. neoformans, with MICs and MLCs that were identical to, or lower than, those of the latter. This contrasts with several reports which indicated that multilamellar liposomal formulations of amphotericin B (in which the phospholipid component consists of DMPC and DMPG in a 7:3 molar ratio) are less active in vitro than free amphotericin B (14, 24). Time-kill studies demonstrated consistent differences between the two formulations, the lower rate of killing with the liposomal amphotericin B preparation suggesting that the antibiotic must dissociate from the phospholipid carrier before producing a fungicidal effect (24).

This investigation is the first to have compared the in vitro fungistatic and fungicidal activities of the four parenteral formulations of amphotericin B that are now available for clinical use. Our results confirm those of several previous investigations which found that liposomal amphotericin B is less active in vitro against C. albicans than is amphotericin B deoxycholate (22, 28). However, an earlier investigation showed that liposomal amphotericin B and free amphotericin B had comparable in vitro fungistatic and fungicidal activities against Aspergillus spp., Candida spp., Fusarium spp., and C. neoformans (1). Our results also support those of a previous report which found that the concentrations of ABCD required to inhibit the growth of Aspergillus spp., Candida spp., and C. neoformans in vitro are sometimes higher than those of amphotericin B deoxycholate (11).

Legrand et al. (16) have demonstrated that differences in structure and lipid composition affect the rate at which amphotericin B is released from lipid-based preparations of this agent. After 1 h of incubation in aqueous solution, the proportion of amphotericin B released from preparations containing 0.05 µg/ml ranged from 20% for liposomal amphotericin B to 75% for ABLC. In other tests, Legrand et al. (16) measured potassium ion release from C. albicans cells during short-term (1 h) incubation with different amphotericin B preparations. Liposomal amphotericin B was about ninefold less active than free amphotericin B against the one strain tested, while ABLC was almost as active as free amphotericin B. In longer-term experiments (24 h of incubation) with the same C. albicans strain, the differences between the preparations were less marked, but liposomal amphotericin B was still twofold less active than free amphotericin B (16). It seems that differences in the rates of release of the agent from the phospholipid carrier could well account for the marked differences in fungistatic and fungicidal activities that we and others have observed among the four amphotericin B preparations.

Comparison of the MICs of nystatin and amphotericin B deoxycholate for the different organisms studied in this work suggests an almost constant relationship between them. For instance, the MIC₅₀s of the two agents for C. albicans, C. glabrata, and C. parapsilosis were 0.25 and 2 µg/ml, while the corresponding values for C. krusei, C. lusitaniae, and C. tropicalis were 0.5 and 4 µg/ml (Table 2). Likewise, the MIC₅₀s of the two agents for A. flavus and A. fumigatus were 1 and 8 µg/ml. This relationship between the susceptibilities of different organisms to amphotericin B and nystatin is not unexpected, given their similar mechanisms of action. Both drugs are believed to bind to ergosterol in fungal cell membranes, causing impairment of membrane barrier function (25).

Our results suggest that organisms which are resistant to one agent might be cross-resistant to the other. There are published reports of C. tropicalis strains which were resistant to both nystatin and amphotericin B (30). However, there are also reports of amphotericin B-resistant strains of C. albicans that were not cross-resistant to nystatin (3). An ergosterol-deficient, nystatin-resistant mutant of C. albicans was shown to be cross-resistant to amphotericin B, but in contrast, an ergosterol-producing, amphotericin B-resistant mutant of the same fungus was not cross-resistant to nystatin (3). These data suggest that the mechanisms of action of the two agents are different in some fungal strains.

Incorporation of antimicrobial drugs into phospholipid carriers has proved most useful for agents that have dose-limiting toxicities. Two such drugs are amphotericin B and nystatin. Lipid-based parenteral formulations of amphotericin B are less toxic than the conventional preparation, and this has enabled higher doses to be used in the treatment of human fungal infections (13). Likewise, liposomal nystatin is much less toxic to animals than is nystatin (19). Initial data from healthy human subjects indicates that parenteral administration of liposomal nystatin, at doses of 2 to 5 mg/kg of body weight, is well tolerated and results in concentrations in serum of 4.8 to 24.1 mg/liter at the end of the infusion (6). Although these levels are higher than the minimal fungicidal concentrations of liposomal nystatin for most of the fungal strains tested in this work, there is a limit to how far the results of in vitro studies can be used to predict drug behavior in human infections.

In conclusion, our work indicates that, although nystatin is less active in vitro than amphotericin B, it does possess useful fungistatic and fungicidal activities against Aspergillus spp., Candida spp., and C. neoformans. Whether the new liposomal formulation of nystatin will have a clinical advantage over amphotericin B cannot be predicted from this in vitro evidence, because there are a large number of other factors, such as concentrations in tissue of the agent at the site of infection, that contribute to clinical outcome. However, initial results from animal models suggest that liposomal nystatin is effective in disseminated Aspergillus infection (9, 29). Liposomal nystatin has also been reported to be active in some human patients who failed to respond to amphotericin B (2). Our findings suggest that further evaluation of liposomal nystatin is justified.