Comparative In Vitro Activities of Caspofungin and Micafungin, Determined Using the Method of the European Committee on Antimicrobial Susceptibility Testing, against Yeast Isolates Obtained in France in 2005-2006

Abstract

The in vitro activities of caspofungin and micafungin against 1,038 yeast isolates have been determined. The caspofungin and micafungin MICs were lower for Candida albicans, Candida glabrata, and Candida tropicalis than for Candida parapsilosis, Candida guilliermondii, and Candida krusei. A clear correlation was seen between the MICs for the two drugs.

Echinocandins are lipopeptide antifungal drugs targeting the fungal cell wall by inhibition of the beta-1-3-glucan synthase (9, 10).

For any given method, the matter of the best technical parameters for antifungal susceptibility testing of echinocandins, particularly for distinction between isolates with low and high MICs, is still debated (5, 21, 23). In this study, we used the broth microdilution reference procedure of the Antifungal Susceptibility Testing Subcommittee of the European Committee on Antimicrobial Susceptibility Testing (AFST-EUCAST) to test a large collection of yeast clinical isolates from France for their in vitro susceptibilities to caspofungin and micafungin. For caspofungin, good agreement between results obtained with the EUCAST method and those obtained with the Clinical and Laboratory Standards Institute (CLSI) technique has been shown (6), and the EUCAST methodology has previously been used to generate in vitro data for this drug (7, 8).

A total of 1,038 yeast isolates, mostly (84%) recovered from blood or sterile sites, consecutively received at the French National Reference Center for Mycoses and Antifungals in 2005-2006, were prospectively analyzed for their in vitro susceptibilities to caspofungin and micafungin. Identification was confirmed using ID32C strips (BioMérieux, Marcy-l'Etoile, France) and ITS1-5.8S-ITS2 sequencing for species other than Candida albicans, Candida glabrata, Candida tropicalis, Candida parapsilosis, and Cryptococcus neoformans. A specific PCR (11) was performed to differentiate Candida dubliniensis from C. albicans.

In vitro susceptibility was determined by following the guidelines of AFST-EUCAST discussion document 7.1 (29). Pure powders of caspofungin and micafungin were used. Microplates were prepared and stored frozen at −20°C. Testing was performed with RPMI-1640 medium supplemented with 2% glucose and adjusted to a final inoculum size of 10⁵ CFU/ml. The final concentrations of the antifungals were 0.015 to 8 μg/ml for both echinocandins. MICs were determined spectrophotometrically after 24 h or 48 h of incubation (depending on the species) at 35°C. The MIC endpoint was defined as 50% or more reduction in growth compared to that in the drug-free well. Two reference strains, Candida krusei ATCC 6258 and C. parapsilosis ATCC 22019, were included in each set of determination.

The MIC₅₀s and MIC₉₀s of the isolates tested were determined for genera for which ≥5 and ≥10 isolates, respectively, were available. MIC distributions were compared by the Mann-Whitney test. Correlation between MIC results for the two echinocandins was assessed by the Pearson correlation coefficient after log₂ transformation.

The in vitro activities of caspofungin and micafungin against all isolates are presented in Table 1. The most susceptible species (MIC₉₀s of ≤1 μg/ml and ≤0.125 μg/ml for caspofungin and micafungin, respectively) were C. albicans, C. glabrata, and C. tropicalis, whereas C. parapsilosis, C. guilliermondii, and C. krusei exhibited higher MICs (MIC₉₀s of ≥2 μg/ml and ≥0.25 μg/ml for caspofungin and micafungin, respectively).

TABLE 1.

In vitro activities of caspofungin and micafungin against 1,038 yeast isolates in France (2005-2006)

Species (nb)	MIC (μg/ml)a of:
	Caspofungin				Micafungin
	Range	MIC50	MIC90	GMIC	Range	MIC50	MIC90	GMIC
C. albicans (404)	0.125-4	0.5	0.5	0.38	0.015-1	0.03	0.06	0.03
C. glabrata (157)	0.25-8	0.5	1	0.51	0.015-8	0.03	0.06	0.03
C. parapsilosis (109)	0.5-4	2	2	1.57	0.25-2	2	2	1.47
C. tropicalis (62)	0.25-2	0.5	1	0.60	0.03-2	0.06	0.125	0.08
C. guilliermondii (27)	0.06-2	2	2	1.32	0.5-2	1	2	1.00
C. krusei (21)	0.5-8	1	2	1.10	0.125-4	0.25	0.25	0.27
C. kefyr (21)	0.125-0.5	0.25	0.5	0.27	0.06-0.5	0.125	0.25	0.12
C. haemulonii (12)	0.25-8	1	2	0.89	0.06-0.25	0.25	0.25	0.18
C. lusitaniae (12)	1-2	1	1	1.06	0.06-0.5	0.25	0.5	0.23
C. fermentati (10)	1-2	1	2	1.41	0.25-1	1	1	0.66
Other Candida spp.c (27)	0.125-2	0.5	1	0.68	0.03-1	0.06	1	0.09
Other Ascomycota yeastsd (13)	0.25-16	1	16	1.38	0.03-16	0.25	16	0.29
C. neoformans (158)	1-16	8	16	9.59	2-16	16	16	11.31
Other Basidiomycota yeastse (5)	8-16	16	ND	ND	8-16	16	ND	ND

^aMICs were determined by the EUCAST method in RPMI medium. High off-scale MICs (>8 μg/ml) were converted to the next higher concentration (16 μg/ml). GMIC, geometric mean MIC; ND, not determined.

^bNumber of isolates.

^cOther Candida spp.: C. dubliniensis (n = 7), C. lipolytica (n = 5), C. famata (n = 2), C. pelliculosa (n = 2), C. rugosa (n = 2), C. utilis (n = 2), and one each of C. inconspicua, C. intermedia, C. africana, C. lambica, C. nivariensis, C. norvegensis, and C. sphaerica.

^dOther Ascomycota yeasts: Saccharomyces cerevisiae (n = 8), Geotrichum spp. (n = 2), and Williopsis saturnus, Kodamaea ohmeri, and a Zygosaccharomyces sp. (n =1 each).

^eOther Basidiomycota yeasts: Trichosporon spp. (n = 3) and Rhodotorula mucilaginosa (n = 2).

Among other Candida spp., less commonly isolated, 90% of the isolates tested (belonging to 17 different species) were inhibited by 2 μg/ml of caspofungin and 1 μg/ml of micafungin. Nevertheless, there was a trend toward higher MICs recorded for both echinocandins against the closely related species Candida fermentati and Candida guilliermondii. Among the other Ascomycota yeast isolates, Geotrichum spp. showed high MICs (>8 μg/ml) for both drugs. Both echinocandins had no activity against C. neoformans, with MIC₉₀s of >8 μg/ml, and similar high MICs were found for the other Basidiomycota yeasts (Trichosporon spp. and Rhodotorula spp.).

Globally, micafungin MICs were significantly lower than caspofungin MICs (P < 0.0001). Nevertheless, a clear correlation (P < 0.0001, Pearson coefficient R² = 0.82) was seen between the MICs for the two drugs, as shown in Table 2.

TABLE 2.

Distribution of micafungin MICs according to caspofungin MICs by use of the EUCAST method in RPMI medium for yeast isolates collected in France

Caspofungin MICa (μg/ml)	No. of isolates with indicated micafungin MICa (μg/ml)
	0.015	0.03	0.06	0.125	0.25	0.5	1	2	4	8	16
0.015
0.03
0.06
0.125	2	9		2	1
0.25	26	147	20	8	2
0.5	22	265	82	20	11	1	1
1	1	24	16	18	21	13	27	28	1
2		1		1	10	7	34	48	2
4						1	2	3	3	4	8
8					1			1	9	19	43
16								1	1	12	59

^aIsolates with MICs of ≥8 μg/ml for both echinocandins were of C. neoformans (n = 125), Trichosporon spp. (n = 3), Rhodotorula mucilaginosa (n = 2), Geotrichum spp. (n = 2), and C. glabrata (n = 1).

Good activities of caspofungin and micafungin against most of the Candida species were found, with C. albicans, C. tropicalis, and C. glabrata being the more susceptible species. Similar susceptibilities to caspofungin were previously reported for isolates in the United States (18, 22, 26, 27) and in Europe (28), as demonstrated by surveys in Spain (7, 8), Germany (13), Denmark (1), and Italy (32). These species were also among the most susceptible to micafungin (16, 22, 25, 30). In contrast, C. parapsilosis and C. guilliermondii showed higher MICs for both echinocandins, as reported in different studies (7, 8, 13, 16, 18, 22, 25-28, 30). A recent animal study showed that although caspofungin treatment significantly reduced renal fungal burden in mice infected by C. parapsilosis or C. guilliermondii, the CFU reduction was 100-fold less than that for C. albicans (4), thereby suggesting intrinsic in vivo reduced susceptibility against the former species. Of note, 2008 Infectious Diseases Society of America guidelines suggested the use of another antifungal class against C. parapsilosis (24). Among less frequently isolated species, C. fermentati, a species phylogenetically closely related to C. guilliermondii but indistinguishable by standard phenotypic criteria (2), exhibited caspofungin MICs similar to those of C. guilliermondii.

Both echinocandins lack in vitro activity against C. neoformans, as previously reported (5, 12, 15, 31), although the reason remains uncertain (10). This lack of in vitro activity is extended to other yeasts belonging to the Basidiomycetes, such as Trichosporon spp. and Rhodotorula spp. Since these yeasts currently emerge in immunocompromised hosts, empirical therapy using an echinocandin in this setting may be inappropriate.

Few studies have compared the in vitro activities of the different echinocandins (19, 22, 25). In some of these studies, MICs were similar for caspofungin and micafungin (19, 25), while in others, such as the present study, micafungin MICs were lower than caspofungin MICs (22). The reason for these differences remains unclear and warrants further investigations. In the present study, the isolates less susceptible to caspofungin were also less susceptible to micafungin and this is in accordance with the results of a recent study (26). Moreover, in several case reports, the clinical failure of echinocandin treatment was associated with elevated caspofungin MICs for the infecting isolates and it was shown that these isolates also had elevated MICs for micafungin and/or anidulafungin (3, 14, 17, 20). Together, these data suggest that there is a “cross-resistance” between caspofungin and micafungin in yeasts.