AAC Accepts, published online ahead of print on 7 May 2012 Antimicrob. Agents Chemother. doi:10.1128/AAC.00355-12 Copyright © 2012, American Society for Microbiology. All Rights Reserved.

Arendrup Page 1

1

Caspofungin Etest susceptibility testing of Candida species: Risk of misclassification of

2

susceptible isolates of C. glabrata and C. krusei when adopting the revised caspofungin

3

CLSI breakpoints.

4

Running Title: Caspofungin Etest and Candida

5

Maiken Cavling Arendrup1*, Michael A.Pfaller2 and the Danish Fungaemia Study Group**.

6 7

**Danish Fungaemia Study Group: Arendrup MC (coordinator)1, Dzajic E3,4, Johansen HK5,

8

Kjældgaard P6, Knudsen JD7, Kristensen L8, Leitz C9, Lemming LE10, Nielsen L11, Olesen B12,

9

Rosenvinge FS13, Røder BL14 and Schønheyder HC15.

10

From: 1Unit of Mycology, Dept. Microbiological Surveillance and research, Statens Serum Institut,

11

Copenhagen, Denmark, 2JMI Laboratories and the University of Iowa, Iowa City, Iowa,

12

USA.3Sydvestjysk Sygehus, Esbjerg, Denmark, 4Vejle Sygehus, Vejle, Denmark, 5Rigshospitalet,

13

Copenhagen University Hospital, Denmark, 6Sygehus Sønderjylland, Sønderborg, Denmark,

14

7

15

9

Hvidovre University Hospital, Hvidovre, Denmark, 8Herning Hospital, Herning, Denmark, Sydvestjysk Sygehus, Viborg, Denmark, 10Skejby Hospital, Aarhus University Hospital, Aarhus,

16

Denmark, 11Herlev University Hospital, Herlev, Denmark , 12Hillerød Hospital, Hillerød, Denmark,

17

13

18

15

19

Key words: Candida, susceptibility testing, Etest, caspofungin

20

*Corresponding author:

Odense University Hospital, Odense, Denmark, 14Slagelse Sygehus, Slagelse, Denmark, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark

Maiken Cavling Arendrup, MD, PhD, Head of Unit of Mycology

21

Unit of Mycology (43/117)

22

Statens Serum Institut

23

Ørestads Boulevard 5

24

DK-2300 Copenhagen S

25

Denmark

26

Email: [email protected] Tel: +45 3268 3223 (Fax: +45 3268 8180)

27 28

Word count: Abstract 70; Text: 1070.

Arendrup Page 2

29

Abstract

30

The purpose of this study was to evaluate the performance of caspofungin Etest and the recently

31

revised CLSI breakpoints. A total of 497 blood isolates were included of which 496 were wild type

32

isolates. 65/496 susceptible isolates (13.1%) were misclassified as intermediate (I) or R. Such

33

misclassifications were most commonly observed for C. krusei (73.1%) and C. glabrata (33.1%).

34

The revised breakpoints cannot be safely adopted for these two species.

35

Arendrup Page 3

36

Text

37

The CLSI breakpoints for the three echinocandins have been revised (16).The motivation behind

38

the change was emerging data suggesting the initial breakpoint defining susceptibility for all

39

Candida species and echinocandins (S: 32 mg/L.

84

The revised CLSI breakpoints were carefully selected in order to provide optimal separation

85

between wild type isolates and isolates with resistance mutations and established either at the

86

epidemiological cut off value (ECV) (C. glabrata) or a single step higher (16). CLSI caspofungin

87

MIC50/ECVs were 0.03/0.125 mg/L for C. albicans, C. glabrata, C. krusei and C. tropicalis (14). The

88

values are, however, one to three dilution steps lower than those found in our study for Etest

89

endpoints (Fig. 1). A comparison of these distributions by species reveals the revised CLSI

90

susceptibility breakpoint bisects the caspofungin Etest endpoint distributions for C. glabrata and C.

91

krusei which thus leads to random classification of wild type isolates as either S or I/R (Fig. 1).

92

One of the strengths for the reference microdilution methods is that growth inhibition is evaluated

93

relative to the growth control for the specific isolates, thereby reducing variation associated with

94

differences in inoculum concentration and growth rate. The automated reading applied in the

95

EUCAST method additionally avoid potential subjectivity in the endpoint reading thereby

96

minimizing variability even further compared to the endpoint reading for agar diffusion methods.

97

However, the Etest MIC50 values reported in this study are in agreement with those reported

98

previously (0.125-0.25 mg/L for C. glabrata and 0.25-1 mg/L for C. krusei) (2, 5, 7, 13, 15, 21) and

99

thus our findings of a high risk of misclassifications will apply to all routine laboratories using

100

caspofungin Etest and CLSI breakpoints. Moreover, in this study all Etest endpoints were read

101

after 24 h of incubation. For C. glabrata a second reading after 48 h is recommended by the

102

manufacturer particularly in cases of weak growth. Previous studies have shown that the Etest

Arendrup Page 5

103

MICs are typically similar or 1 dilution step higher after 48h and thus a second reading would

104

further increase the risk of mis-classifications (5).

105

In conclusion, this study illustrates the caveats associated with the adoption of reference

106

breakpoints for commercial methods when MIC distributions do not exactly mirror one another. In

107

the case of caspofungin Etest, the revised CLSI breakpoints can be safely adopted for C. albicans,

108

C. dubliniensis, C. parapsilosis and C. tropicalis, but not for C. glabrata and C. krusei. According to

109

this and our previous studies an Etest susceptibility breakpoint of S: 32

1

1 2 19 9

7

3 1

MIC50 (mg/L)

Range (mg/L)