Susceptibility Testing of Anidulafungin and Voriconazole Alone and in Combination against Conidia and Hyphae of Aspergillus spp. under Hypoxic Conditions

ABSTRACT MICs and fractional inhibitory concentrations were evaluated for anidulafungin and voriconazole alone and in combination against conidia and hyphae under hypoxic (1% oxygen-5% CO2-94% nitrogen) conditions against 31 Aspergillus isolates. Anidulafungin exhibited excellent activity against conidia and hyphae of Aspergillus spp. The visual reading of the MIC for anidulafungin was optimal under hypoxic conditions.

Filamentous fungal pathogens are recognized as a major and increasing source of infection in immunocompromised hosts (3). The most common species causing disease are Aspergillus fumigatus (90%) followed by Aspergillus flavus, Aspergillus niger, and Aspergillus terreus (3). Due to the high mortality rate from mold infections and the limited efficacies of the current agents, the use of combination therapy is an interesting option in the treatment of invasive aspergillosis (11,15). Since echinocandins and triazoles are different classes of antifungal drugs with different modes of action, the use of this two-drug combination has recently received much attention in medical mycology (2,8,12,14).
Anidulafungin is a novel cyclic lipopeptide antifungal agent of the echinocandin class, which acts via inhibition of 1,3-␤-Dglucan synthesis (18). The drug end point for the echinocandins is the minimum effective concentration (MEC) (1,5). However, determination of MEC is difficult and labor intensive and requires expertise. Warn et al. (19) demonstrated that end point reading for caspofungin and micafungin under hypoxic conditions (1% oxygen) was superior and suggests reading MICs rather than MECs. Therefore, we investigated the in vitro activity of anidulafungin alone and in combination with the azole voriconazole against conidia and hyphae of Aspergillus spp. under hypoxic conditions.
Ten clinical isolates of A. fumigatus, nine of A. terreus, and six each of A. flavus and A. niger were investigated. Strains were obtained from patients suffering from invasive aspergillosis.
The MICs of anidulafungin (kindly provided by Pfizer, Vienna, Austria) and voriconazole (kindly provided by Pfizer, Vienna, Austria) for Aspergillus spp. were tested according to the proposed Antifungal Susceptibility Testing Subcommittee of the European Committee on Antibiotic Susceptibility Testing broth microdilution method (9) but under hypoxic conditions. MICs for hyphae were evaluated by the method of Lass-Flörl et al. (10). Briefly, 100 l of conidial solutions were added to 96-well plates (Costar, Vienna, Austria) and incubated at 30°C for 16 to 22 h to allow the formation of hyphae. The outgrowth of hyphal length (50 to 70 m) was determined by using an inverted microscope. Wells were washed and refilled with 100 l RPMI 2% glucose, and antifungal agents were added and incubated at 37°C for 48 h (10). The microdilution plates were incubated in a gas-tight jar with an oxygen concentration of 1% according to the protocol of Warn et al. (19), and MIC end points were read at 100% inhibition. Also, the metabolic activity of drug-treated hyphae was determined from their ability to reduce the tetrazolium compound 2,3-bis(2methoxy-4-nitro-5-sulfophenyl)-2 H-tetrazolium-5-carboxanilide (XTT) as described elsewhere (4).
For selected experiments, the plates were incubated in ambient air conditions (20% oxygen). For voriconazole, MICs were read visually with no growth end point, and for anidulafungin, the MEC instead of the MIC was defined microscopically (6).
The synergy tests were evaluated by using MIC end points of each drug. The fractional inhibitory concentration (FIC) of each drug for an individual isolate was calculated as the ratio of the concentration of the drug in combination that achieves the MIC end point to the MIC of the drug alone using that end point. FIC index (FICI) values were interpreted as follows: FICI of Յ0.5, synergistic; FICI of Ͼ0.5 to Յ4, indifferent; and FICI of Ͼ4, antagonistic (7). All tests were performed twice and in duplicate.
In this study we determined the in vitro antifungal activity of anidulafungin either alone or in combination with voriconazole against conidia (Table 1) and hyphae ( Table 2) by reading the MIC on the basis of a complete growth inhibition end point under hypoxic conditions. We found that the growth of Aspergillus spp. under these conditions in the presence of anidulafungin was almost totally inhibited, and the end point was clear and plain to define. The isolates tested were susceptible to anidulafungin and voriconazole, displaying lower MICs (Յ0.06 g/ml) for anidulafungin than those found for voriconazole (Յ2 g/ml) under hypoxic conditions, as shown in Tables 1 and 2. A subset of Aspergillus isolates (n ϭ 10) tested under ambi-ent air conditions demonstrated that MECs and MICs for conidia (n ϭ 10) and hyphae (n ϭ 10) of anidulafungin and voriconazole correlated to 93% and 90% (data not shown) with the MICs tested under hypoxic conditions. Hence, the hypoxic condition had no influence on the MICs of voriconazole. MECs and MICs of anidulafungin were also similar within the various testing conditions. The onset of invasive Aspergillus infection is associated with the appearance of hyphae (6). Anidulafungin displayed excellent results against hyphae (Table 2), with MICs as low as those found for conidia (Table 1). Voriconazole also exhibited excellent in vitro activity against hyphae of Aspergillus spp. (Table  2) but at slightly higher concentrations than those needed to inhibit conidial growth (Table 1), which is in accordance with a previously published report (13). For hyphae, comparison of the visually determined end points with the results of the XTT method revealed that 80.7% of the visually determined MICs corresponded to a reduction in metabolic activity (data not shown).
As shown in Table 1, the in vitro interaction of anidulafungin with voriconazole in combination for a conidial suspension of Aspergillus spp. exerted partly synergy and a lack of antagonism. Similar results were obtained by others (14). Yet this report is the first to demonstrate the interaction of anidula-fungin and voriconazole against hyphae of Aspergillus. Results for the majority of isolates were indifferent, whereas results for 4 of the 27 tested isolates showed antagonism (Table 2). Presently we lack in vivo data for this drug combination. The use of animal models to evaluate the usability of anidulafungin combined with voriconazole for combination therapy is warranted.
The fact that oxygen levels in wounds and deep-tissue foci are much lower (Ͻ1% oxygen) than in healthy tissues (16) may support testing under hypoxic conditions. Under hypoxic conditions, similar to those that might occur in tissue infected with Aspergillus (17), the end point for anidulafungin was clear and easy to define. The MEC and MIC were found to be similar, yet the MIC was easier to read.