In Vitro Pharmacodynamics of Anidulafungin and Caspofungin against Candida glabrata Isolates, Including Strains with Decreased Caspofungin Susceptibility

Abstract

The activities of anidulafungin and caspofungin against Candida glabrata were evaluated. MICs, 50% inhibitory concentrations (IC₅₀ values), and IC₉₀ values for anidulafungin were lower than those for caspofungin for 16 of 18 strains tested. Anidulafungin has potent in vitro activity against C. glabrata that is maintained against isolates with elevated caspofungin MICs.

Infections caused by Candida glabrata have been reported to be increasing in frequency (19, 30) and are associated with high mortality rates for the elderly, immunocompromised patients, and patients in intensive care units (2, 11, 27). Treatment may be complicated by the variable susceptibility of C. glabrata to fluconazole and by reduced response rates to other azole antifungals in breakthrough infections (2, 11, 22, 26, 29).

The echinocandins have emerged as an effective treatment strategy for invasive candidiasis. Although susceptibility breakpoints are not currently established, broth microdilution studies have demonstrated relatively good activity for each member of this class against Candida isolates, including non-C. albicans species (13, 28). However, case reports describing clinical failures of caspofungin associated with reduced in vitro activity against C. albicans, C. glabrata, and C. krusei have begun to emerge (6, 7, 9, 10, 14, 21). It is currently unknown if other echinocandins would maintain potency at clinically relevant exposures in the face of diminished caspofungin activity. The objective of this study was to compare the in vitro activities and pharmacodynamics of anidulafungin and caspofungin against Candida glabrata isolates over a range of clinically achievable concentrations.

(Part of this work was presented previously [N. P. Wiederhold, J. R. Graybill, L. K. Najvar, and D. S. Burgess, Abstr. 45th Intersci. Conf. Antimicrob. Agents Chemother., abstr. M-2160, 2005].)

Eighteen Candida glabrata isolates from the University of Texas Health Science Center at San Antonio were tested. Stock solutions of anidulafungin (Vicuron Pharmaceuticals, King of Prussia, PA) and caspofungin (Merck & Co., Inc., Whitehouse Station, NJ) were prepared by dissolving drug powders in dimethyl sulfoxide and water, respectively.

Microdilution broth susceptibility testing was performed in duplicate according to the CLSI M27-A2 method in RPMI growth medium buffered with 0.165 M 4-morpholinepropanesulfonic acid (15). The MIC₂ was defined as the lowest concentration of anidulafungin or caspofungin that caused a significant decrease in turbidity (≥50%) compared to that of the growth control, and the MIC₀ was defined as the lowest concentration resulting in no visual growth. The minimum fungicidal concentration (MFC) was measured as previously described (5).

The 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) colorimetric assay was performed in triplicate as previously reported (12). Final anidulafungin concentrations ranged from 0.015 to 32 μg/ml, and those of caspofungin ranged from 0.06 to 128 μg/ml. The absorbance was read at 492 nm, and readings were converted to percent absorbance, with values for growth control wells set at 100% and those for medium control wells set at 0%. XTT reduction assay data were fit to a four-parameter inhibitory sigmoid model, using computer curve fitting software (Prism 4; GraphPad Software, Inc., San Diego, CA), to derive 50% inhibitory concentration (IC₅₀) and IC₉₀ values. The goodness of fit for each isolate/drug was assessed by the R² value and the standard error of the IC₅₀ value. The IC₅₀ and IC₉₀ values were not determined if minimal or no reduction in formazan absorbance was observed with increasing drug concentrations.

Time-kill studies were performed in duplicate on isolates with caspofungin MICs of ≥8 μg/ml, using previously described methods (8). The anidulafungin and caspofungin concentrations tested were 0.5, 1.0, 4.0, 8.0, and 16 μg/ml. These are within the range achieved clinically for both agents (3, 4, 25). Fungicidal activity was defined as a ≥3-log₁₀ (99.9%) reduction in CFU from the starting inoculum.

Anidulafungin MICs ranged from 0.125 to 4 μg/ml and were at least two dilutions lower than those of caspofungin (range, 1 to 64 μg/ml) for 16 of the 18 isolates tested, including the three isolates with elevated caspofungin MICs (Table 1). Similarly, the MFCs for anidulafungin (range, 0.25 to 8 μg/ml) were ≥2 dilutions lower than those of caspofungin (1 to ≥128 μg/ml) for 14 of the 18 isolates tested. Anidulafungin remained fungicidal against the three strains with elevated caspofungin MFCs (≥64 μg/ml).

TABLE 1.

MIC, MFC, IC₅₀, and IC₉₀ values for anidulafungin and caspofungin against C. glabrata isolates

Isolate	Anidulafungin value			Caspofungin value
	MIC2/MIC0/MFC	IC50/IC90	R2	MIC2/MIC0/ MFC	IC50/IC90	R2
1	0.25/0.5/0.5	0.063/0.075	0.97	1/2/4	0.81/0.83	0.91
2	0.125/0.25/0.25	0.11/0.18	0.93	1/1/2	0.52/0.60	0.96
3	0.125/0.25/0.25	0.039/0.11	0.86	1/2/4	0.55/0.61	0.95
4	0.125/0.25/0.25	0.086/0.18	0.91	1/2/2	0.91/1.1	0.89
5	0.125/0.25/0.25	0.15/0.65	0.83	1/2/2	0.64/0.85	0.89
6	1/2/4	0.65/2.4	0.96	2/2/128	1.8/23	0.73
7	0.125/0.25/0.5	0.092/0.15	0.96	1/1/2	0.57/0.67	0.95
8	0.25/0.25/0.5	0.049/0.16	0.94	1/1/2	0.90/1.0	0.82
9	0.25/0.5/0.5	0.13/0.15	0.88	1/1/2	0.94/1.0	0.91
10	0.125/0.25/1	0.11/0.19	0.81	1/1/2	0.86/1.2	0.91
11	0.125/0.25/4	0.071/0.12	0.98	1/1/1	0.91/0.98	0.86
12	0.125/0.25/0.5	0.11/0.12	0.93	1/1/4	0.84/0.94	0.95
13	0.25/0.25/1	0.053/0.38	0.89	1/1/2	0.74/1.0	0.89
14	0.125/0.125/1	0.062/0.21	0.92	1/1/2	0.90/0.99	0.90
15	0.125/0.125/1	0.11/0.13	0.83	1/1/4	0.83/0.93	0.94
16	1/2/4	0.17/7.8	0.78	4/8/64	NDa	ND
17	2/4/8	0.83/5.3	0.93	64/64/>128	ND	ND
18	0.5/4/8	2.4/3.6	0.90	2/64/64	ND	ND

^aND, IC₅₀ and IC₉₀ values were not determined because of minimal or no reduction in formazan absorbance with increasing drug concentrations.

The potency of anidulafungin was greater than that of caspofungin, as evidenced by a lower IC₅₀ value for each strain evaluated (Table 1). Similarly, the IC₉₀ values for anidulafungin were also lower than those for caspofungin against all isolates. Against the isolates with caspofungin MICs of ≥8 μg/ml, only the highest concentration of caspofungin tested (128 μg/ml) resulted in a >90% reduction in viability, while anidulafungin maintained its potency (IC₅₀ range, 0.17 to 2.4 μg/ml; IC₉₀ range, 3.6 to 7.8 μg/ml).

These results are supported by data from time-kill studies (Fig. 1). Against C. glabrata isolate 16, anidulafungin at 1 μg/ml resulted in a >3-log₁₀ CFU/ml decrease in the starting inoculum at 24 h, compared to a 0.43-log₁₀ CFU/ml reduction for caspofungin at 16 μg/ml. Anidulafungin at 4 μg/ml was fungicidal against isolate 18, in contrast to the 0.54-log₁₀ CFU/ml increase observed with caspofungin at 16 μg/ml. Minimal activity was also observed for caspofungin against C. glabrata isolate 17 at the highest concentration tested (0.65-log₁₀ increase in CFU/ml at 16 μg/ml). Although anidulafungin was not fungicidal against this isolate, concentrations of ≥4 μg/ml did result in reductions in colony counts (≥1.74-log₁₀ CFU/ml decrease).

FIG. 1.

Dose-response curves (A, D, and G) and time-kill plots (B, C, E, F, H, and I) for anidulafungin and caspofungin against C. glabrata isolates with caspofungin MICs of ≥8 μg/ml. (A, B, and C) C. glabrata isolate 16; (D, E, and F) C. glabrata isolate 17; (G, H, and I) C. glabrata isolate 18. Symbols in dose-response curves (A, D, and G) show means ± standard errors for experiments performed in triplicate with anidulafungin (•) and caspofungin (○). Curves were generated by fitting XTT data to a four-parameter inhibitory sigmoid model. Symbols in time-kill plots (B, C, E, F, H, and I) show the mean values for log₁₀ CFU/ml versus time for anidulafungin (B, E, and H) and caspofungin (C, F, and I) against C. glabrata isolates with caspofungin MICs of ≥8 μg/ml. Drug concentrations: ▪, control; ▴, 0.5 μg/ml; ⧫, 1 μg/ml; •, 4 μg/ml; ▾, 8 μg/ml; and *, 16 μg/ml. Symbols represent the means ± standard errors for experiments performed in duplicate for each drug. E_max values represent maximum log₁₀ reductions in CFU/ml for anidulafungin and caspofungin against each isolate at 24 h.

Surveillance studies have reported relatively good activities for the echinocandins compared to other antifungals against Candida species, including non-C. albicans and fluconazole-resistant isolates (1, 5, 16, 17, 24). Against the isolates of C. glabrata tested for this study, anidulafungin had greater in vitro potency than did caspofungin, as evidenced by lower MIC and MFC values. The enhanced potency of anidulafungin against C. glabrata isolates is also supported by pharmacodynamic data from XTT and time-kill assays and was maintained against C. glabrata isolates with elevated caspofungin MICs.

While no strong correlation exists between susceptibility data and clinical success in the treatment of invasive fungal infections, in vitro and in vivo data suggest that antifungal MICs may be predictive of responses to therapy (7, 9). Furthermore, antifungal resistance is associated with clinical failure (9, 18, 20, 23). Recent reports suggested that this may also be true for caspofungin in the setting of elevated MICs (6, 7, 9, 10, 14, 21). Due to the difficulty in treating invasive candidiasis that is unresponsive to prior therapy, further studies investigating the utility of anidulafungin in this setting are warranted.