J Antimicrob Chemother 2019; 74: 2295–2302

doi:10.1093/jac/dkz178 Advance Access publication 12 May 2019

Antifungal activity and killing kinetics of anidulafungin, caspofungin

and amphotericin B against Candida auris

Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019

Catiana Dudiuk1,2, Indira Berrio3,4, Florencia Leonardelli1,2, Soraya Morales-Lopez5,6, Laura Theill1,
Daiana Macedo1,2, José Yesid-Rodriguez7, Soraya Salcedo8, Adriana Marin8, Soledad Gamarra1 and
Guillermo Garcia-Effron1,2*

1
Laboratorio de Micologı́a y Diagnóstico Molecular, Cátedra de Parasitologı́a y Micologı́a, Facultad de Bioquı́mica y Ciencias Biológicas,
Universidad Nacional del Litoral, Santa Fe, Argentina; 2Consejo Nacional de Investigaciones Cientı́ficas y Tecnológicas (CONICET), CCT,
Santa Fe, Argentina; 3Medical and Experimental Mycology Group, Corporación para Investigaciones Biológicas (CIB), Medellı́n,
Colombia; 4Hospital general de Medellin ‘Luz Castro de Gutiérrez’ ESE, Medellı́n, Colombia; 5Universidad de Santander, Facultad de
Ciencias de la Salud, Grupo de Investigación CIENCIA UDES, Valledupar, Colombia; 6Laboratorios Nancy Flórez Garcı́a S.A.S., Valledupar,
Colombia; 7Centro de Investigaciones Microbiológicas del Cesar, CIMCE, Valledupar, Colombia; 8Clı́nica General del Norte, Barranquilla,
Colombia

*Corresponding author. Tel: !54 342 4575209; Fax: !54 342 4575216; E-mail: ggarcia@unl.edu.ar

Received 2 January 2019; returned 19 February 2019; revised 8 March 2019; accepted 2 April 2019

Background: Candida auris is an emerging MDR pathogen. It shows reduced susceptibility to azole drugs and, in
some strains, high amphotericin B MICs have been described. For these reasons, echinocandins were proposed
as first-line treatment for C. auris infections. However, information on how echinocandins and amphotericin B
act against this species is lacking.
Objectives: Our aim was to establish the killing kinetics of anidulafungin, caspofungin and amphotericin B
against C. auris by time–kill methodology and to determine if these antifungals behave as fungicidal or fungistat-
ic agents against this species.
Methods: The susceptibility of 50 C. auris strains was studied. Nine strains were selected (based on echinocandin
MICs) to be further studied. Minimal fungicidal concentrations, in vitro dose–response and time–kill patterns
were determined.
Results: Echinocandins showed lower MIC values than amphotericin B (geometric mean of 0.12 and 0.94 mg/L,
respectively). Anidulafungin and caspofungin showed no fungicidal activity at any concentration (maximum log
decreases in cfu/mL between 1.34 and 2.22). On the other hand, amphotericin B showed fungicidal activity, but
at high concentrations (2.00 mg/L). In addition, the tested polyene was faster than echinocandins at killing
50% of the initial inoculum (0.92 versus .8.00 h, respectively).
Conclusions: Amphotericin B was the only agent regarded as fungicidal against C. auris. Moreover, C. auris
should be considered tolerant to caspofungin and anidulafungin considering that their MFC:MIC ratios were
mostly 32 and that after 6 h of incubation the starting inoculum was not reduced in .90%.

Introduction reduced susceptibility to the other triazoles including the newest

Candida auris is an emerging and multiresistant yeast pathogen
that is becoming an important agent of hospital outbreaks.1–5
Isolates are grouped into different clades representing different
geographical regions that coincide with different susceptibility pat-
terns.3 This species was considered intrinsically resistant to flucon-
azole since initial reports showed high (.64 mg/L) MIC values for
all the studied strains.6–8 Now and after studying a larger and
more diverse collection of strains, it has been concluded that flu-
conazole resistance can be acquired.3,6,7 C. auris also shows
(e.g. isavuconazole).7 Moreover, amphotericin B susceptibility
cannot be anticipated since a wide range of MIC values has been
obtained.7,9–11 This scenario led the US CDC, the IDSA and other
medical associations to propose echinocandins as first-line treat-
ment.12–14
It is known that each antifungal drug works differently depend-
ing on the studied species.15,16 Thus, echinocandins behave as
fungicidal agents against the most common Candida spp.,17–20
but as fungistatic agents against species showing intrinsic reduced

C The Author(s) 2019. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
V
For permissions, please email: journals.permissions@oup.com.
2295
Dudiuk et al.
echinocandin susceptibilities (e.g. Candida parapsilosis sensu lato,
Candida lusitaniae and Candida guilliermondii).15,21–23 Despite the
importance of C. auris, information about how echinocandin drugs
and amphotericin B act against this species is lacking. The aims of
this study were to establish the killing kinetics of anidulafungin,
caspofungin and amphotericin B against C. auris by time–kill meth-
odology and to determine whether these antifungals behave as
fungicidal or fungistatic agents.
Materials and methods
Isolates
A total of 50 C. auris strains isolated between 2015 and 2016 in different
Colombian cities (Medellin, Valledupar and Barranquilla) were studied. All
the strains were isolated from individual patients with proven invasive fun-
gal disease (candidaemia). Additionally, Candida parapsilosis sensu stricto
ATCC 22019 and Candida krusei ATCC 6258 were used as susceptibility test-
ing controls. C. auris isolates were identified using a multiplex classical PCR
technique published by our group24 and confirmed by sequencing of the
5.8S ribosomal gene and the adjacent internal transcribed spacer 1 (ITS1)
and ITS2 regions.25
Antifungal susceptibility testing (MIC determination),
minimal fungicidal concentration (MFC) determination
and FKS1 sequencing
The susceptibilities of the 50 C. auris strains to amphotericin B, caspofungin
and anidulafungin were evaluated following the CLSI documents M27 and
M60.26,27 Antifungal drugs were purchased from Sigma–Aldrich (Argentina)
or obtained from their manufacturers as standard powders. According to
the echinocandin MIC values, 9 of the 50 strains were selected and divided
into three groups following the proposed epidemiological cut-off values
(ECOFFs):6 (i) WT–WT group (three strains), WT for both tested echinocan-
dins (caspofungin and anidulafungin MICs ,0.25 mg/L); (ii) NWT–WT group
(four strains), non-WT for caspofungin and WT for anidulafungin; and (iii)
NWT–NWT group (two strains), non-WT for both tested echinocandins
(caspofungin and anidulafungin MICs 0.25 mg/L). Although there are no
established caspofungin ECOFFs, we decided to use the anidulafungin
ECOFFs for both echinocandins with the sole intention of classifying the iso-
lates into groups and to be able to analyse their behaviour when tested
against the studied drugs. MFC values of the same nine C. auris strains were
obtained for the three antifungals as described previously.15,21 Succinctly,
MFC experiments were performed by inoculating the complete volume
(200 lL) of the visually clear wells (where no growth was seen in microtitre
wells) onto Sabouraud dextrose agar (SDA) plates. The starting inoculum
for MFC determinations was 5%105 cells/mL (obtained by counting in a
haemocytometer). MFCs were considered as the lowest concentrations of
drug that killed 99.9% of the starting inoculum (,10 colonies/plate). Strains
included in the NWT–WT and NWT–NWT groups were subjected to FKS1
gene sequencing (hot spot regions) as described by Chowdhary et al.7
Antifungal carry-over
Yeast inocula of 2%105 cells/mL (obtained by using a haemocytometer)
were prepared in sterile water. These inocula were diluted 1:2 in 2% RPMI-
1640 containing the tested drugs. These dilutions were made in order to
yield a final fungal suspension of 1%105 cells/mL.28 Antifungal carry-over
was immediately evaluated by inoculating SDA plates with the described
inoculum. After 24 h of incubation, the obtained cfu at each tested drug
concentration was compared with the cfu of the drug-free counterpart.
Antifungal carry-over was defined as a 25% reduction in cfu when com-
pared with drug-free control.28
2296
Time–kill curves and in vitro dose–response studies
Time–kill curve assays were carried out with the nine selected C. auris
strains by using microdilution plates and RPMI-1640 buffered with MOPS as
described previously.28 The isolates were grown for 24 h at 35 C on SDA
plates. Caspofungin, anidulafungin and amphotericin B concentrations
tested ranged from 0.12 to 8.00 mg/L. Inoculum suspensions of 1%105
cells/mL were used. Microdilution plates were incubated at 35 C for 2, 4, 6,
Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019
8, 10, 24 and 48 h. Aliquots of 50 lL of each well were serially diluted (10-
fold) in sterile water, spread onto SDA plates and incubated at 35 C for 48 h
to determine the number of cfu/mL. Time–kill curve assays were conducted
at least in duplicate and on separate days to ensure reproducibility of the
results.15,20 The same nine C. auris strains were used for in vitro dose–
response evaluation. Starting inocula of 105 cells/mL were exposed for 24 h
to different antifungal concentrations (0.12–8.00 mg/L) in microdilution
plates. Subsequently, 50 lL aliquots of each well were serially diluted and
plated to count the remaining cfu/mL.
Mathematical model, data analysis and definitions
The MIC data are expressed as geometric means (GMs) of three experi-
ments performed on different days. The off-scale MICs and MFCs were con-
verted to the next concentration up or down in order to be included in the
analysis. The killing kinetics were studied by fitting the data to an exponen-
tial equation: Nt " No % e#Kt (Nt, number of viable cells at time t; No, number
of cells at the beginning of the assay; K, killing rate; and t, incubation
time).29 This equation was linearized by applying natural logarithms. K val-
ues were used to compare the antifungal killing rates (killing when K is
negative, growing when it is positive). The eight timepoints were reduced to
one K value for each killing curve (mean values). R2 values 0.8 were con-
sidered as the limit for the goodness of fit. Mean times to reach 50% and
99.9% (fungicidal endpoint) reduction of the initial inoculum were derived
from the killing kinetics equation (t50"0.30103/K and t99.9"3/K).30 On the
other hand, in vitro potencies of the different drugs were evaluated using a
sigmoidal dose–effect model by comparing the dose needed to reduce the
initial inoculum by 50% (IC50). The significance of the differences in killing
kinetics between drugs, concentration of drugs and isolates were evaluated
by analysis of variance (ANOVA) and pairwise comparisons. P,0.05 was
considered significant. Antifungals were considered to behave as static
or cidal agents following the published definitions.31,32 Briefly, a drug
was considered as a fungicidal agent when its MFC:MIC ratio was 4, but as
fungistatic when its MFC:MIC ratio was between .4 and ,32. Moreover,
a strain was considered tolerant to a particular drug when its MFC:MIC
ratio was 32.31,32
Results
MICs and MFCs
Echinocandin MICs for the 50 C. auris strains were distributed
among a wide range of values (0.03–0.25 and 0.03–1.00 mg/L for
anidulafungin and caspofungin, respectively). Anidulafungin
showed slightly lower MIC50 and MIC90 than caspofungin (0.12 ver-
sus 0.25 mg/L and 0.25 mg/L versus 1.00 mg/L, respectively), but
similar GMs (0.12 for both drugs). On the other hand, amphotericin B
MIC values were spread over three dilutions ranging from 0.50 to
2.00 mg/L (MIC GM"0.94 mg/L, and MIC50 and MIC90"1.00 mg/L).
Overall, following the ECOFFs of Arendrup et al.,6 all but four strains
were considered WT for anidulafungin (8%). For caspofungin there
are no published ECOFFs; however, 19 strains showed caspofungin
MICs .0.25 mg/L. On the other hand, only three strains (6%) would
be deemed non-WT for amphotericin B (MICs .1.00 mg/L).6
Following the described ECOFF-based classification,6 27 of the
C. auris strains were included in the WT–WT group, 19 in the
Anidulafungin, caspofungin and amphotericin B against C. auris JAC
Table 1. Activities of caspofungin, anidulafungin and amphotericin B against isolates of C. auris (MICs, MFCs and MFC:MIC ratios)

MIC (mg/L) MFC (mg/L) MFC:MIC ratiob

Strain Pattern (CAS–AFG)a CAS AFG AMB CAS AFG AMB CAS AFG AMB

LMDM-1162 WT–WT 0.06 0.12 0.50 .8.00 .8.00 4.00 267 (T) 133 (T) 4 (FC)

Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019

LMDM-1242 WT–WT 0.06 0.03 1.00 .8.00 1.00 4.00 267 (T) 33(T) 4 (FC)
LMDM-1245 WT–WT 0.06 0.06 1.00 .8.00 .8.00 4.00 267 (T) 267 (T) 4 (FC)
LMDM-1198 NWT–WT 1.00 0.12 2.00 .8.00 4.00 2.00 16 (FS) 33 (T) 1 (FC)
LMDM-1205 NWT–WT 0.25 0.06 1.00 .8.00 .8.00 2.00 64 (T) 267 (T) 2 (FC)
LMDM-1213 NWT–WT 0.50 0.12 1.00 .8.00 4.00 2.00 32 (T) 33 (T) 2 (FC)
LMDM-1168 NWT–WT 0.50 0.12 1.00 .8.00 .8.00 2.00 32 (T) 133 (T) 2 (FC)
LMDM-1217 NWT–NWT 0.50 0.25 1.00 .8.00 4.00 2.00 32 (T) 16 (FS) 2 (FC)
LMDM-1234 NWT–NWT 1.00 0.50 1.00 .8.00 .8.00 4.00 16 (FS) 32 (T) 4 (FC)

CAS, caspofungin; AFG, anidulafungin; AMB, amphotericin B.

a
Echinocandin susceptibility patterns. Strains were classified as WT or NWT based on published ECOFFs.6
b
MFC:MIC ratios were obtained by using the 24 h MIC data and considering the MFC values .8.00 mg/L as 16.00 mg/L (T, tolerant; FC, fungicidal; and
FS, fungistatic).

NWT–WT set and 4 in the NWT–NWT group. None of the 23 strains
classified as NWT–WT or NWT–NWT showed FKS1 hot spot muta-
tions. Out of these strains, nine were selected (three WT–WT, four
NWT–WT and two NWT–NWT strains) for further study (Table 1).
Caspofungin MFCs were .8 mg/L for all the tested strains
(independent of their MICs). MFC values for anidulafungin were al-
most as high as those obtained for the other echinocandin (MFC
GM"7.41 mg/L). However, anidulafungin was able to reach the
99.9% inhibition threshold in four strains, but at very high concen-
trations (1 and 4 mg/L). Three of these strains were considered
WT for anidulafungin including the strain with the lowest MIC
of this drug in the entire collection (LMDM-1242). Regarding
amphotericin B, this polyene was able to kill 99.9% of the starting
inoculum of the nine studied strains, with MFC values ranging from
2.00 to 4.00 mg/L (GM"2.72 mg/L).
Considering the MFC:MIC ratios,31,32 the only agent able to be
regarded as fungicidal against all the tested C. auris strains
(MFC:MIC ratios 4) was amphotericin B. On the other hand, these
ratios for caspofungin and anidulafungin were higher (arithmetic
mean"110.33 and 105.22, respectively). Thus, C. auris should be
considered tolerant to the studied echinocandins (MFC:MIC ratios
were 32 for all but two strains for caspofungin and for all but one
for anidulafungin) (Table 1). It has to be pointed out that these
ratios were calculated considering a hypothetical value of 16 mg/L
for strains whose MFCs were .8.00 mg/L. Thus, the three MFC:MIC
ratios regarded as fungistatic (.4, but ,32) are more likely to be
due to the high MIC values than to the low MFC values.
Antifungal carry-over, in vitro dose–response studies
and time–kill curves
No antifungal carry-over was observed for any of the used drugs or
any of the C. auris strains. The lowest IC50 arithmetic means were
obtained for anidulafungin followed by that achieved by caspofun-
gin and by amphotericin B (0.22, 0.63 and 0.74 mg/L, respectively)
(Table 2) (P"0.005 for anidulafungin versus caspofungin, P"0.558
for amphotericin B versus caspofungin and P"0.0001 for anidula-
fungin versus amphotericin B). However, caspofungin was not able
to decrease the starting inoculum by half after 24 h (no IC50 values
were obtained) for almost half of the studied C. auris strains
(Table 2 and Figure 1). Regarding anidulafungin, there were no
statistically significant differences between the IC50 values
obtained for the strains classified as WT or non-WT (arithmetic
mean"0.244 versus 0.208, respectively; P"403).
Figure 2 depicts the obtained killing curves. For echinocandins,
two different slopes can be distinguished. The maximum
decreases in cfu/mL were obtained at the 6 h timepoint. In these
first hours, the killing activities were concentration independent for
the two studied echinocandins (negative values of K that were not
substantially modified by drug concentration augmentation).
After 6 h of incubation, regrowth was observed (positive values
of K). This regrowth was concentration dependent for anidulafun-
gin, but concentration independent for caspofungin. Concerning
the studied polyene, inflection points were also evident. However,
remarkable differences were seen when compared with echino-
candins. Lethality rates (K values) were totally concentration
dependent (Figure 3).
Differences in the maximum log decreases in cfu/mL between
the different antifungal classes were observed. For echinocandins,
the maximum log decreases in cfu/mL never surpassed 3 log
(99.9% of the initial inoculum) since these log reductions were be-
tween 1.34 and 2.22 for caspofungin and between 1.57 and 2.15
for anidulafungin (reached at 0.50–1.00 mg/L caspofungin and
between 0.25 and 0.50 mg/L anidulafungin). Thus, there was no
fungicidal activity for the tested antifungals of this class at any
concentration and at any timepoint. On the other hand, the max-
imum log decreases for amphotericin B were achieved at concen-
trations 2.00 mg/L and were widely superior to 3 log decreases
in cfu/mL (4.41–5.18 log decreases). K values were negative
(indicating killing) for all the drugs against all the strains, with the
exception of caspofungin against LMDM-1242, LMDM-1213 and
LMDM-1234 (Table 2). Despite this clear negative trend, K values
differed between the echinocandins and amphotericin B. The aver-
age K for anidulafungin and caspofungin was 5.7 and 12.8 times
less negative than for amphotericin B, respectively (average K
arithmetic means for all nine C. auris strains were as follows:

2297
Dudiuk et al.

Table 2. Killing activity, killing kinetics and time to achieve 50% and 99.9% reduction from the starting inoculum of caspofungin, anidulafungin and
amphotericin B against C. auris

IC50 (mg/L) Average K (killing kinetics) (h#1) t50 (h)b t99.9 (h)b

Strain Pattern (CAS–AFG)a CAS AFG AMB CAS AFG AMB CAS AFG AMB CAS AFG AMB

Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019

LMDM-1162 WT–WT 0.475 0.213 0.481 #0.032 #0.038 #0.665 9.4 7.9 0.5 93.8 78.9 4.5
LMDM-1242 WT–WT 0.867 0.277 1.162 0.006 #0.074 #0.310 NK 4.1 1.0 NK 40.5 9.7
LMDM-1245 WT–WT 0.320 0.170 0.560 #0.068 #0.074 #0.303 4.4 4.1 2.9 44.1 40.5 9.9
LMDM-1198 NWT–WT 0.338 0.240 0.430 #0.043 #0.048 #0.469 7.0 6.3 0.6 69.8 62.5 6.4
LMDM-1205 NWT–WT NR 0.125 1.041 #0.018 #0.0464 #0.397 16.7 6.5 0.8 166.7 64.7 7.6
LMDM-1213 NWT–WT 1.173 0.216 0.901 0.029 #0.048 #0.772 NK 6.3 0.4 NK 62.5 3.9
LMDM-1168 NWT–WT NR 0.213 0.425 #0.033 #0.056 #0.498 9.1 5.4 0.6 90.9 53.6 6.0
LMDM-1217 NWT–NWT NR 0.200 0.630 #0.060 #0.059 #0.256 5.0 5.1 1.2 50.0 50.8 11.7
LMDM-1234 NWT–NWT NR 0.288 1.060 0.042 #0.005 #0.913 NK 60.2 0.3 NK 600.0 3.3

CAS, caspofungin; AFG, anidulafungin; AMB, amphotericin B; NR, 50% inhibition never reached.
a
Echinocandin susceptibility patterns based on published ECOFFs.6
b
t50 and t99.9 were obtained using the average K of each strain [NK, no killing (K values were positive)].

LMDM-1162 (WT-WT) 120 LMDM-1168 (NWT-WT)

120
CAS
100 100 AFG

80 AMB
Survival (%)

80
Survival (%)

60 60

40 40

20 20

0 0
–3 –2 –1 0 1 2 3 –3 –2 –1 0 1 2 3
Log2 drug concentration (mg/L) Log2 drug concentration (mg/L)

120 LMDM-1217 (NWT-NWT)

100
Survival (%)

80

60

40

20

0
–3 –2 –1 0 1 2 3
Log2 drug concentration (mg/L)

Figure 1. Archetypal in vitro dose–effect curves comparing the doses of caspofungin, anidulafungin and amphotericin B needed to reduce the initial
inoculum by 50% (IC50s) in WT–WT, NWT–WT and NWT–NWT groups (WT–WT, WT for both tested echinocandins; NWT–WT, non-WT for caspofungin
and WT for anidulafungin; and NWT–NWT, non-WT for both tested echinocandins).

anidulafungin, #0.096; caspofungin, #0.043; and amphotericin B,
#0.55). Moreover, the average K values for echinocandins were
very close to zero (at some concentrations they were .0, and ,0
for other drug concentrations) (Figure 3). The mean times needed
to kill 50% of the initial inoculum (t50) were 0.92, 11.77 and 8.60 h
for amphotericin B, anidulafungin and caspofungin, respectively
(Table 2). However, it has to be noted that caspofungin was not
able to reach this percentage of killing in one-third of the tested
C. auris strains. These strains were classified as WT–WT, NWT–WT
and NWT–NWT (one each), showing that the killing capability of

2298
Anidulafungin, caspofungin and amphotericin B against C. auris JAC
Caspofungin Anidulafungin Amphotericin B
WT-WT WT-WT WT-WT
8
7 7 7
6 6 6 0.12

Log cfu/mL
5 5 0.25
Log cfu/mL

Log cfu/mL
4 4 0.5
4

Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019

1
3 3 3 2
2 2 2 4
1 1 8
1
0 0 0
0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48
Time (h) Time (h) Time (h)

NWT-WT 8 NWT-WT 8 NWT-WT

8
7 7
7
6 6 0.12
6
Log cfu/mL

Log cfu/mL
0.25
Log cfu/mL

5 5
5 0.5
4 4 4
1
3 3 3 2
2 2 2 4
1 1 1 8
0 0 0
0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48
Time (h) Time (h) Time (h)
NWT-NWT
NWT-NWT NWT-NWT 8
8
8 7
7
7 6
6 0.12

Log cfu/mL
6
Log cfu/mL

5 0.25
Log cfu/mL

5 5
4 4 0.5
4 1
3 3
3 2
2 2 2
4
1 1 1 8
0 0 0
0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48
Time (h) Time (h) Time (h)

Figure 2. Representative time–kill plots following drug exposure for strains of each of the three echinocandin susceptibility groups (WT–WT, WT for
both tested echinocandins; NWT–WT, non-WT for caspofungin and WT for anidulafungin; and NWT–NWT, non-WT for both tested echinocandins).
Error bars are omitted to reduce graphic complexity.

caspofungin seems not to be related to its MIC values. For all the
tested strains and drugs, the minimum t50 values were obtained at
4% the strain’s MIC values. On the other hand, the time needed
to reach a hypothetical fungicidal endpoint (99.9% inhibition) was
.48 h for echinocandins for most of the strains (Table 2). As
expected, t99.9 values for amphotericin B were lower than for
caspofungin and anidulafungin (arithmetic mean"9.13 versus
.48 h) (P,0.01). However, these values were not related to
amphotericin B MIC values (as for echinocandins) and were
reached between 2% and 4% the MIC (Table 2).
Discussion
To the best of our knowledge, this is the first work where the killing
kinetics of caspofungin, anidulafungin and amphotericin B against
C. auris were compared. We found several similarities between our
results and those obtained for other Candida species such as the
taxonomically related C. lusitaniae and Candida species showing
an intrinsic reduced echinocandin susceptibility phenotype such as
the C. guilliermondii and C. parapsilosis species complex.22,29,33 We
found that the behaviour of amphotericin B against C. auris was
close to that observed for species traditionally considered resist-
ant to amphotericin B or susceptible to developing secondary re-
sistance to this drug (C. lusitaniae and C. guilliermondii).33–36 The
rate of killing for C. auris by amphotericin B is concentration de-
pendent, as described for C. lusitaniae and C. guilliermondii.
However, we found that C. auris amphotericin B K values are not
strain dependent (smaller range of K values).15,22 Moreover, in
amphotericin B-susceptible species (e.g. C. albicans), the 99.9%
killing is reached rapidly (t99.9 2 h) and at the strains’ MIC val-
ues.15,22 In contrast, the 3 log reductions in cfu/mL were
achieved 3–4 times more slowly for C. auris than for amphoteri-
cin B-susceptible species and at amphotericin B concentrations
4% their MICs. Notably, C. auris t99.9 values were also higher than
those obtained for C. lusitaniae16 and for C. guilliermondii.22
Adding to these data, we also show that most of the strains
needed .2.00 mg/L amphotericin B to inhibit 99.9% of the start-
ing inoculum. This concentration of drug is higher than the
amphotericin B concentration needed to reach the 3 log de-
crease in cfu/mL in C. guilliermondii, but similar to the polyene
concentration needed to kill 99.9% of the starting inoculum of
amphotericin B-resistant C. lusitaniae strains.15,22,34

2299
Dudiuk et al.

(a) LMDM-1242 (WT-WT) Turning to echinocandin drugs, several reports demonstrated

CAS
that these antifungals behave as fungicidal agents against the
AFG
0.5 majority of Candida species.23,37 However, neither caspofungin nor
AMB
anidulafungin showed fungicidal activity (99.9% killing) against
0.0
our C. auris strains. The data presented here resemble the killing
–0.5
kinetic patterns observed for echinocandins against intrinsic

Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019

reduced echinocandin susceptibility (e.g. C. guilliermondii and
K (cfu/mL/h)

–1.0 C. parapsilosis sensu stricto)22,34 and taxonomically related species

(C. lusitaniae).33,34
–1.5 There are several objective ways of evaluating these killing pat-
terns. For bacteria, it has long been known that the MFC:MIC ratio
–2.0
can be used to establish whether or not a drug is cidal. It was
–2.5
established that a drug should be considered cidal, static or toler-
ant if these ratios are 4, between 4 and 32 or 32, respective-
–3.0 ly.32,38 Moreover, older reports defined tolerance as a
–3 –2 –1 0 1 2 3 phenomenon in which organisms that are normally killed by a cidal
Log2 drug concentration (mg/L)
agent are just inhibited39 and that clinically important tolerance
(b) LMDM-1213 (NWT-WT) should be identified by time–kill curves and has to be defined as
CAS
90% inhibition of the starting inoculum after 6 h.40 Furthermore,
0.5 AFG
AMB
tolerance phenotypes in bacterial pathogens are described mainly
0.0
for cell wall-acting antibacterial agents.32,39–42 Considering these
definitions and our results, C. auris should be considered as a toler-
–0.5 ant species to anidulafungin and caspofungin since all the
described characteristics for a tolerant species are fulfilled (includ-
K (cfu/mL/h)

–1.0 ing the fact that echinocandins are cell wall-acting antifungals).
This study provides evidence of the non-fungicidal activity of
–1.5 caspofungin and anidulafungin, and the extremely high ampho-
tericin B concentration needed to reach 99.9% inhibition. However,
–2.0 it has been known for some time that this type of study is far from
reflecting a real in vivo situation where, for example, the drug
–2.5
undergoes metabolism and elimination, the inoculum size is vari-
–3.0
able and diffusion/distribution restrictions exist in different
–3 –2 –1 0 1 2 3 tissues.11 Moreover, pharmacokinetic/pharmacodynamic animal
Log2 drug concentration (mg/L) models suggested that echinocandins are the most efficacious
drug class against C. auris.43 Despite these limitations and consid-
(c) LMDM-1217 (NWT-NWT) CAS
ering the few therapeutic options available to treat C. auris infec-
0.5 AFG
tions, it is clear that further studies are needed to uncover the
AMB
mechanism involved in these phenotypes and to determine if
0.0
these in vitro observations have any clinical implications.
–0.5

Acknowledgements
K (cfu/mL/h)

–1.0
We thank Dr Stella Vaira (Mathematic Department, Facultad de
–1.5 Bioquı́mica y Ciencias Biológicas, Universidad Nacional del Litoral) for
helpful suggestions regarding statistics.
–2.0

–2.5
Funding
This study was supported by the Science, Technology and Productive
–3.0
–3 –2 –1 0 1 2 3
Innovation Ministry (MinCyT; Argentina) grant PICT2016/1985 (to
Log2 drug concentration (mg/L) G. G.-E.). F. L. and D. M. each have a fellowship from CONICET
(Argentina). D. M. has received an Overseas Scholarship from the British
Figure 3. Relationship of K values (calculated from regression lines) and Society for Antimicrobial Chemotherapy (Overseas Scholarship
caspofungin, anidulafungin and amphotericin B concentrations for WT– BSAC-2017-0022). C. D. has a postdoctoral fellowship from CONICET.
WT (a), NWT–WT (b) and NWT–NWT (c) groups (WT–WT, WT for both
tested echinocandins; NWT–WT, non-WT for caspofungin and WT for
anidulafungin; and NWT–NWT, non-WT for both tested echinocandins).
Echinocandin regression lines are virtually parallel to the concentration Transparency declarations
axis (values near 0). None to declare.

2300
Anidulafungin, caspofungin and amphotericin B against C. auris
References
1 Ben-Ami R, Berman J, Novikov A et al. Multidrug-resistant Candida haemu-
lonii and C. auris, Tel Aviv, Israel. Emerg Infect Dis 2017; 23. doi:
10.3201/eid2302.161486.
2 Lee WG, Shin JH, Uh Y et al. First three reported cases of nosocomial funge-
mia caused by Candida auris. J Clin Microbiol 2011; 49: 3139–42.
3 Lockhart SR, Etienne KA, Vallabhaneni S et al. Simultaneous emergence of
multidrug-resistant Candida auris on 3 continents confirmed by whole-
genome sequencing and epidemiological analyses. Clin Infect Dis 2017; 64:
134–40.
4 Morales-Lopez SE, Parra-Giraldo CM, Ceballos GA et al. Invasive infections
with multidrug-resistant yeast Candida auris, Colombia. Emerg Infect Dis
2017; 23: 162–4.
5 Schelenz S, Hagen F, Rhodes JL et al. First hospital outbreak of the globally
emerging Candida auris in a European hospital. Antimicrob Resist Infect
Control 2016; 5: 35.
6 Arendrup MC, Prakash A, Meletiadis J et al. Comparison of EUCAST and CLSI
reference microdilution MICs of eight antifungal compounds for Candida auris
and associated tentative epidemiological cutoff values. Antimicrob Agents
Chemother 2017; 61: e00485-17.
7 Chowdhary A, Prakash A, Sharma C et al. A multicentre study of antifungal
susceptibility patterns among 350 Candida auris isolates (2009–17) in India:
role of the ERG11 and FKS1 genes in azole and echinocandin resistance.
J Antimicrob Chemother 2018; 73: 891–9.
8 Kathuria S, Singh PK, Sharma C et al. Multidrug-resistant Candida
auris misidentified as Candida haemulonii: characterization by matrix-assisted
laser desorption ionization-time of flight mass spectrometry and DNA
sequencing and its antifungal susceptibility profile variability by Vitek 2, CLSI
broth microdilution, and Etest method. J Clin Microbiol 2015; 53: 1823–30.
9 Calvo B, Melo AS, Perozo-Mena A et al. First report of Candida auris in
America: clinical and microbiological aspects of 18 episodes of candidemia.
J Infect 2016; 73: 369–74.
10 Magobo RE, Corcoran C, Seetharam S et al. Candida auris-associated can-
didemia, South Africa. Emerg Infect Dis 2014; 20: 1250–1.
11 Shin JH, Kim MN, Jang SJ et al. Detection of amphotericin B resistance in
Candida haemulonii and closely related species by use of the Etest, Vitek-2
yeast susceptibility system, and CLSI and EUCAST broth microdilution meth-
ods. J Clin Microbiol 2012; 50: 1852–5.
12 CDC. Candida auris: Information for Laboratorians and Health
Professionals. 2018. https://www.cdc.gov/fungal/candida-auris/recommenda
tions.html? CDC_AA_refVal"https%3A%2F%2Fwww.cdc.gov%2Ffungal
%2Fdiseases%2Fcandidiasis%2Frecomm.
13 IDSA. Candida auris Clinical Update—September 2017. https://www.cdc.
gov/fungal/candida-auris/c-auris-alert-09-17.html.
14 Sears D, Schwartz BS. Candida auris: an emerging multidrug-resistant
pathogen. Int J Infect Dis 2017; 63: 95–8.
15 Canton E, Peman J, Gobernado M et al. Patterns of amphotericin B killing
kinetics against seven Candida species. Antimicrob Agents Chemother 2004;
48: 2477–82.
16 Pfaller MA, Sheehan DJ, Rex JH. Determination of fungicidal activities
against yeasts and molds: lessons learned from bactericidal testing and the
need for standardization. Clin Microbiol Rev 2004; 17: 268–80.
17 Barchiesi F, Spreghini E, Tomassetti S et al. Comparison of the fungicidal
activities of caspofungin and amphotericin B against Candida glabrata.
Antimicrob Agents Chemother 2005; 49: 4989–92.
18 Ernst EJ, Roling EE, Petzold CR et al. In vitro activity of micafungin (FK-
463) against Candida spp.: microdilution, time–kill, and postantifungal-effect
studies. Antimicrob Agents Chemother 2002; 46: 3846–53.
JAC
19 Gil-Alonso S, Jauregizar N, Canton E et al. Comparison of the in vitro activ-
ity of echinocandins against Candida albicans, Candida dubliniensis, and
Candida africana by time–kill curves. Diagn Microbiol Infect Dis 2015; 82:
57–61.
20 Gil-Alonso S, Jauregizar N, Canton E et al. In vitro fungicidal activities of
anidulafungin, caspofungin, and micafungin against Candida glabrata,
Candida bracarensis, and Candida nivariensis evaluated by time–kill studies.
Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019
Antimicrob Agents Chemother 2015; 59: 3615–8.
21 Canton E, Peman J, Viudes A et al. Minimum fungicidal concentrations of
amphotericin B for bloodstream Candida species. Diagn Microbiol Infect Dis
2003; 45: 203–6.
22 Canton E, Peman J, Sastre M et al. Killing kinetics of caspofungin, micafun-
gin, and amphotericin B against Candida guilliermondii. Antimicrob Agents
Chemother 2006; 50: 2829–32.
23 Canton E, Peman J, Valentin A et al. In vitro activities of echinocandins
against Candida krusei determined by three methods: MIC and minimal
fungicidal concentration measurements and time–kill studies. Antimicrob
Agents Chemother 2009; 53: 3108–11.
24 Theill L, Dudiuk C, Morales-Lopez S et al. Single-tube classical PCR for
Candida auris and Candida haemulonii identification. Rev Iberoam Micol 2018;
35: 110.
25 White TJ, Bruns TD, Lee SB et al. Amplification and direct sequencing of
fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH,
Sninsky JJ et al., eds. PCR Protocols: A Guide to Methods and Applications.
San Diego, CA, USA: Academic Press, Inc., 1990; 315–22.
26 Clinical and Laboratory Standards Institute. Performance Standards for
Antifungal Susceptibility Testing of Yeasts—First Edition: M60. CLSI, Wayne, PA,
USA, 2017.
27 Clinical and Laboratory Standards Institute. Reference Method for Broth
Dilution Antifungal Susceptibility Testing of Yeasts—Fourth Edition: M27. CLSI,
Wayne, PA, USA, 2017.
28 Klepser ME, Ernst EJ, Lewis RE et al. Influence of test conditions on anti-
fungal time–kill curve results: proposal for standardized methods. Antimicrob
Agents Chemother 1998; 42: 1207–12.
29 Canton E, Espinel-Ingroff A, Peman J et al. In vitro fungicidal activities of
echinocandins against Candida metapsilosis, C. orthopsilosis, and C. parapsilo-
sis evaluated by time–kill studies. Antimicrob Agents Chemother 2010; 54:
2194–7.
30 Canton E, Peman J. [Antifungal time–kill curves.]. Rev Iberoam Micol
1999; 16: 82–5.
31 Pankey GA, Sabath LD. Clinical relevance of bacteriostatic versus bacteri-
cidal mechanisms of action in the treatment of Gram-positive bacterial infec-
tions. Clin Infect Dis 2004; 38: 864–70.
32 French GL. Bactericidal agents in the treatment of MRSA infections—
the potential role of daptomycin. J Antimicrob Chemother 2006; 58:
1107–17.
33 Canton E, Peman J, Hervas D et al. Examination of the in vitro fungicidal
activity of echinocandins against Candida lusitaniae by time–killing methods.
J Antimicrob Chemother 2013; 68: 864–8.
34 Di BG, Spedicato I, Picciani C et al. In vitro pharmacodynamic
characteristics of amphotericin B, caspofungin, fluconazole, and vori-
conazole against bloodstream isolates of infrequent Candida species
from patients with hematologic malignancies. Antimicrob Agents
Chemother 2004; 48: 4453–6.
35 Colombo AL, Junior JNA, Guinea J. Emerging multidrug-resistant Candida
species. Curr Opin Infect Dis 2017; 30: 528–38.
36 Asner SA, Giulieri S, Diezi M et al. Acquired multidrug antifungal resistance
in Candida lusitaniae during therapy. Antimicrob Agents Chemother 2015; 59:
7715–22.
2301
Dudiuk et al.
37 Clancy CJ, Huang H, Cheng S et al. Characterizing the effects of caspofun-
gin on Candida albicans, Candida parapsilosis, and Candida glabrata isolates
by simultaneous time–kill and postantifungal-effect experiments. Antimicrob
Agents Chemother 2006; 50: 2569–72.
38 Sader HS, Fritsche TR, Jones RN. Daptomycin bactericidal activity and cor-
relation between disk and broth microdilution method results in testing of
Staphylococcus aureus strains with decreased susceptibility to vancomycin.
Antimicrob Agents Chemother 2006; 50: 2330–6.
39 Tuomanen E, Durack DT, Tomasz A. Antibiotic tolerance among
clinical isolates of bacteria. Antimicrob Agents Chemother 1986; 30:
521–7.
2302
40 May J, Shannon K, King A et al. Glycopeptide tolerance in Staphylococcus
aureus. J Antimicrob Chemother 1998; 42: 189–97.
41 Singh M, Matsuo M, Sasaki T et al. In vitro tolerance of drug-naive
Staphylococcus aureus strain FDA209P to vancomycin. Antimicrob Agents
Chemother 2017; 61: e01154-16.
42 Gefen O, Balaban NQ. The importance of being persistent: heterogeneity
of bacterial populations under antibiotic stress. FEMS Microbiol Rev 2009; 33:
Downloaded from https://academic.oup.com/jac/article-abstract/74/8/2295/5488498 by Nottingham Trent University user on 29 July 2019
704–17.
43 Lepak AJ, Zhao M, Berkow EL et al. Pharmacodynamic optimization for
treatment of invasive Candida auris infection. Antimicrob Agents Chemother
2017; 61: e00791-17.