Decreased Killing Activity of Micafungin Against Candida guilliermondii,Candida lusitaniae,and Candida kefyr in the Presence of Human Serum

Abstract

Currently, echinocandins are first-line drugs for treatment of invasive candidiasis. However, data on how serum influences killing activity of echinocandins against uncommon Candida species are limited. Therefore, the killing activity of micafungin in RPMI-1640 and in 50% serum was compared against Candida guilliermondii, Candida lusitaniae, and Candida kefyr. Minimum inhibitory concentration (MIC) ranges in RPMI-1640 were 0.5–1, 0.12–0.25, and 0.06–0.12 mg/L, respectively. In 50% serum, MICs increased 32- to 256-fold. In RPMI-1640 ≥ 0.25, ≥4, and 32 mg/L micafungin was fungicidal against all four C. kefyr (≤4.04 hours), two of three C. lusitaniae (≤16.10 hours), and two of three C. guilliermondii (≤12.30 hours), respectively. In 50% serum, all three species grew at ≤4 mg/L. Micafungin at 16–32 mg/L was fungicidal against all C. kefyr isolates (≤3.03 hours) and at 32 mg/L was fungistatic against one of three C. lusitaniae isolates. Two C. lusitaniae isolates and all three C. guilliermondii grew at all tested concentrations. Adding human serum to susceptibility test media drew attention to loss of fungicidal or fungistatic activity of micafungin in the presence of serum proteins, which is not predicted by MICs in case of C. kefyr and C. lusitaniae in RPMI-1640. Our results strongly suggest that micafungin and probably other echinocandins should be used with caution against rare Candida species.

Introduction

Although five Candida species (Candida albicans, Candida glabrata, Candida parapsilosis sensu stricto, Candida tropicalis, and Candida krusei) account for ≥95% of all candidemia, less common other species (Candida guilliermondii, Candida lusitaniae, and Candida kefyr), may cause problems in the treatment of candidemias or other forms of invasive candidiasis, especially among cancer and leukemia patients.^1–4 Optimal treatment against these species is not well defined as they often show decreased susceptibility to different antifungal classes, including echinocandins.^1,5,6

Currently, echinocandins such as micafungin are first-line drugs for treatment of invasive candidiasis.⁷ Echinocandins are highly protein bound with 0.2–3% free drug, that is, the serum and tissue concentrations may be low.⁸ Decreased killing activity of echinocandins in 50% serum against medically important Candida species was confirmed.^9–12 However, data on how serum influences killing activity of echinocandins against uncommon Candida species are limited.^9–11,13 To extend our knowledge about the in vitro activity of echinocandins against uncommon Candida species, the killing activity of micafungin in RPMI-1640 and in RPMI-1640 plus 50% serum (50% serum) was compared with C. guilliermondii, C. lusitaniae, and C. kefyr.

Materials and Methods

Bloodstream-derived clinical isolates (Table 1) were identified using MICRONAUT-Candida and API ID32C methods.¹⁴ All strains were isolated before antifungal administration. Micafungin pure powder was kindly provided by Astellas. Minimum inhibitory concentrations (MICs) in RPMI-1640 and in 50% serum (serum from a human male, type AB; Sigma, Budapest, Hungary) were determined using the standard broth macrodilution method. Final concentration ranged between 0.03 and 32 mg/L. MIC values were read visually after 24 hours using the partial inhibition criterion.¹⁵

Table 1.

Minimum Inhibitory Concentrations of Micafungin Against Candida kefyr, Candida lusitaniae, and Candida guilliermondii Clinical Isolates in RPMI-1640 With (50% Serum) and Without 50% Serum (RPMI)

	MIC (mg/L)
Clinical isolates	RPMI	50% serum	Ratio
C. kefyr 4273	0.06	16	256
C. kefyr 5243	0.06	8	128
C. kefyr 7194	0.12	8	64
C. kefyr 8003	0.06	8	128
C. lusitaniae 582	0.25	16	64
C. lusitaniae 3834	0.12	16	128
C. lusitaniae 7849	0.12	16	128
C. guilliermondii 5465	1	32	32
C. guilliermondii 5540	0.5	32	64
C. guilliermondii 21060	1	32	32

MIC, minimum inhibitory concentration.

Time kill studies were performed in both media at 1, 4, 16, and 32 mg/L at least twice. In case of C. kefyr 0.25 mg/L micafungin was also tested. The starting inocula were 3.5–5 × 10⁵ CFU/ml. Aliquots of 100 μl were removed at 0, 4, 8, 12, 24, and 48 hours, serially diluted 10-fold, plated onto Sabouraud agar and incubated at 35°C for 48 hours. The same method was used to test growth kinetics.^9–12,16,17

Fungicidal activity of micafungin was defined as 99.9% or higher reduction in viable cell count.^9–13,16,17 Killing kinetics were analyzed in both media (RPMI-1640 with and without serum) as described previously.^11,12 Briefly, an exponential equation was fitted to the mean data at each time point: N_t = N₀ × e^−kt, where N_t is the number of viable yeasts at time t, N₀ is the number of viable yeasts in the initial inoculum, k is the killing rate, and t is the incubation time; reducing time points of each killing curve to one value (k). Positive k values indicate killing, negative values indicate growth. The goodness of fit was assessed by the r² value (r² > 0.8). The mean times to achieve the fungicidal endpoint (T99.9 = 3/k) were calculated from the k values for each isolate and concentration in both media.^11,12

Killing kinetics among isolates and concentrations in either RPMI-1640 or 50% serum were compared using one-way ANOVA with Tukey's post testing; growth rates and the efficacy of the same micafungin concentrations in different media were compared by T-test (with Welch's correction, where appropriate).^11,12

Results

C. guilliermondii isolates were susceptible to micafungin in RPMI-1640 (Table 1).¹⁸ No breakpoint is available for C. kefyr and C. lusitaniae. However, MICs of isolates of these two species were not higher than the epidemiological cut off values (0.12 and 0.5 mg/L for C. kefyr and C. lusitaniae, respectively).¹⁹ In 50% serum, MICs increased 32- to 256-fold. All C. kefyr isolates and C. guilliermondii isolate 5540 grew significantly better in RPMI-1640 than RPMI-1640 plus serum (p < 0.05).

Representative time-kill plots are shown in Fig. 1. In RPMI-1640, killing exerted by micafungin was rapid against C. kefyr; mean times to achieve 99.9% growth reduction were shorter than 4.04 hours for all isolates (Table 2). Contrastingly, 0.25, 1, and 4 mg/L micafungin did not produce any killing against C. kefyr isolates in serum (k values were negative and did not differ significantly from those of the drug-free controls) (Fig. 2 and Table 3). Higher concentrations (16 and 32 mg/L), however, produced fast (≤3.03 hours) killing in 50% serum as well.

FIG. 1.

Representative time–kill curves against Candida kefyr, Candida lusitaniae and Candida guilliermondii in RPMI-1640 with (MIC_SERUM) and without 50% serum (MIC_RPMI).

FIG. 2.

Killing rates of micafungin and the corresponding adjusted regression lines (dashed lines) against four C. kefyr bloodstream isolates in RPMI-1640 with (serum) and without 50% serum (RPMI). Positive and negative k values indicate the decrease and increase, respectively, in viable cell numbers.

Table 2.

Time to Reach 99.9% (T99.9) Growth Reduction at the Indicated Concentrations of Micafungin in RPMI-1640 With (50% Serum) and Without 50% Serum (RPMI)

		Time (hours)
Isolate	Media	0.25 mg/L	1 mg/L	4 mg/L	16 mg/L	32 mg/L
C. kefyr 4273	RPMI	3.48	3.49	3.22	2.43	3.04
	50% serum	Growth	Growth	Growth	NA	3.03
C. kefyr 5243	RPMI	2.25	2.05	2.16	2.39	2.58
	50% serum	Growth	Growth	Growth	NA	2.43
C. kefyr 7194	RPMI	2.83	1.99	2.25	2.52	2.39
	50% serum	Growth	Growth	Growth	2.44	2
C. kefyr 8003	RPMI	3.83	4.04	3.68	3.63	3.88
	50% serum	Growth	Growth	Growth	2.83	2.71
C. lusitaniae 582	RPMI	—	NA	NA	NA	NA
	50% serum	—	Growth	Growth	Growth	Growth
C. lusitaniae 3834	RPMI	—	15.27	12.53	16.1	11.17
	50% serum	—	Growth	Growth	Growth	NA
C. lusitaniae 7849	RPMI	—	NA	10.56	9.35	8.96
	50% serum	—	Growth	Growth	Growth	Growth
C. guilliermondii 5465	RPMI	—	NA	NA	NA	12.3
	50% serum	—	Growth	Growth	Growth	Growth
C. guilliermondii 5540	RPMI	—	NA	NA	NA	NA
	50% serum	—	Growth	Growth	Growth	Growth
C. guilliermondii 21060	RPMI	—	NA	NA	NA	10.7
	50% serum	—	Growth	Growth	Growth	Growth

—, not determined; NA, not achieved.

Table 3.

Maximum Changes in log CFU/ml Compared to the Starting Inoculum in Time-Kill Studies in RPMI-1640 With (50% Serum) and Without 50% Serum (RPMI)

		Maximum log decreases in CFU in time-killing experiments at different micafungin concentrations (mg/L)
Isolate	Media	0.25	1	4	16	32
C. kefyr 4273	RPMI	−3.3	−3.3	−3.3	−3.3	−3.3
	50% serum	Growth	Growth	Growth	−2.48	−3.3
C. kefyr 5243	RPMI	−3.78	−3.78	−3.78	−3.78	−3.78
	50% serum	Growth	Growth	Growth	−2.95	−3.78
C. kefyr 7194	RPMI	−3.78	−3.78	−3.78	−3.78	−3.78
	50% serum	Growth	Growth	Growth	−3.08	−3.78
C. kefyr 8003	RPMI	−3.78	−3.78	−3.78	−3.78	−3.78
	50% serum	Growth	Growth	Growth	−2.56^a	−3.78
C. lusitaniae 582	RPMI	—	−2.08	−2.12	−2.19	−2.08
	50% serum	—	Growth	Growth	−0.16^a	−0.96^a
C. lusitaniae 3834	RPMI	—	−3.78	−3.26	−3.78	−3.78
	50% serum	—	Growth	Growth	−0.19^a	−2
C. lusitaniae 7849	RPMI	—	−2.56^a	−3.78	−3.78	−3.78
	50% serum	—	Growth	−0.23^a	−0.94^a	−1.78^a
C. guilliermondii 5465	RPMI	—	−0.29^a	−1.07^a	−2.44^a	−3.38
	50% serum	—	Growth	Growth	Growth	Growth
C. guilliermondii 5540	RPMI	—	−1.22^a	−2.95^a	−2.86	−2.90^a
	50% serum	—	Growth	Growth	Growth	−0.05^a
C. guilliermondii 21060	RPMI	—	−0.52^a	−1.65^a	−2.71	−3.16
	50% serum	—	Growth	Growth	Growth	Growth

Regrowth occurred.

CFU, colony-forming unit.

Micafungin at 4, 16, and 32 mg/L was fungicidal against two of three C. lusitaniae isolates in RPMI-1640 (Tables 2 and 3); the killing activity of micafungin was strain dependent, but concentration independent (i.e., k values did not differ significantly) (Fig. 3). In 50% serum, micafungin produced only transient colony-forming unit (CFU) decreases even at 16–32 mg/L against C. lusitaniae isolates (except isolate 3834 at 32 mg/L) (Table 3). Positive k value (0.2369 1/h, indicating killing) was noticed only at 32 mg/L in case of isolate 3834 (Fig. 3).

FIG. 3.

Killing rates of micafungin and the corresponding adjusted regression lines (dashed lines) against three–three C. lusitaniae (left) and C. guilliermondii (right) bloodstream isolates in RPMI-1640 (RPMI) and RPMI-1640 plus 50% serum (serum). Positive and negative k values indicate the decrease and increase, respectively, in viable cell numbers.

Although micafungin was fungicidal at 32 mg/L against C. guilliermondii isolates 5465 and 21060 in RPMI-1640, the inhibitory effect was only transient (regrowth occurred) at lower concentrations (1, 4, and 16 mg/L) (Table 3). In RPMI-1640, k values at 16 and 32 mg/L were strain dependent, but concentration independent (Fig. 3). In 50% serum, killing curves for C. guilliermondii at any concentrations were similar to controls (k values were always negative, indicating growth) (Figs. 1 and 3).

Discussion

It was previously shown that C. guilliermondii is intrinsically less susceptible to echinocandins, whereas wild-type C. kefyr and C. lusitaniae MIC values are close to those obtained with innately echinocandin-susceptible C. albicans or C. glabrata.^6,19 However, mortality rates caused by C. guilliermondii, C. lusitaniae, and C. kefyr among seriously ill patients are extremely high (53–83%), suggesting that currently used therapeutic approaches, including echinocandin therapy, are insufficient.^1–4 Furthermore, infections caused by C. lusitaniae and C. kefyr have become increasingly prevalent in parallel with the continuous increase in echinocandin use.^1,6,20

The CLSI introduced clinical break-points for the five most important Candida species (C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, and C. krusei), but not for uncommon (rare) Candida species (with the exception of C. guilliermondii).^15,18 However, it is not taken into account that RPMI-1640 is a serum-free medium, thus in case of the highly protein-bound echinocandins, their in vitro potency against these Candida species may be overestimated in spite of the very low MIC values in standard susceptibility tests.^6,18 Epidemiological cutoff values for the species against which official breakpoints has not yet been published are also based on the standard susceptibility testing using RPMI-1640.^18,19

Serum-based testing of echinocandin susceptibility, which mimics the in vivo conditions occurring in the blood or tissues more closely, has some advantages.²¹ First, in vivo efficacy of micafungin showed better correlation with MICs by serum-based than by standard RPMI-1640 susceptibility tests.²² Second, 50% serum may help to differentiate between wild-type isolates and isolates harboring resistance mutations.^6,21 Third, serum eliminates the paradoxically decreased killing activity at high concentration of echinocandins in RPMI-1640.^12,18,21,23

Killing activities in this study were found to be markedly lower than against C. albicans⁹ in the presence of serum, that is, no killing by 4 mg/L micafungin; even higher concentrations proved to be consistently fungicidal only against C. kefyr. Consequently, reported median peak and trough plasma levels (15.15 and 4.85 mg/L, respectively) produced by the elevated 150 mg micafungin daily doses,²⁴ predict inadequate eradication and increased risk of breakthrough infections¹ by these three species. Concordantly, only weak eradication capacity of echinocandins was reported in neutropenic animal models against these species even after 7–10 days echinocandin therapy (i.e., internal organs were not sterilized by echinocandins).^25–27 In addition, inadequate serum echinocandin levels may facilitate development of acquired resistance to echinocandins.⁶

In summary, adding human serum to susceptibility test media drew attention to loss of killing activity in the presence of serum proteins, which is not predicted by standard MICs in case of C. kefyr and C. lusitaniae in RPMI-1640. Even elevated therapeutic micafungin doses (150 mg/day) do not produce serum drug levels sufficient for reliable fungicidal or fungistatic activity against any of the three studied species, which may explain the observed high probability of therapeutic failure in infections with these species. This issue is especially important in case of patients with neutropenia or other immunosuppression. Although serum-based susceptibility testing has not yet been standardized, our results strongly suggest that micafungin and probably other echinocandins should be used with caution against rare Candida species.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Jung

D.S.

, Farmakiotis

, Jiang

, Tarrand

J.J.

, and Kontoyiannis

D.P.

. 2015. Uncommon Candida species fungemia among cancer patients, Houston, Texas, USA. Emerg. Infect. Dis., 21:1942–1950.

Lortholary

, Renaudat

, Sitbon

, Madec

, Denoeud-Ndam

, Wolff

, Fontanet

, Bretagne

, Dromer

; French Mycosis Study Group. 2014. Worrisome trends in incidence and mortality of candidemia in intensive care units (Paris area, 2002–2010). Intensive Care Med., 40:1303–1312.

Wang

, Farmakiotis

, Yang

, McCue

D.A.

, Kantarjian

H.M.

, Kontoyiannis

D.P.

, and Mathisen

M.S.

. 2015. The ever-evolving landscape of candidaemia in patients with acute leukaemia: non-susceptibility to caspofungin and multidrug resistance are associated with increased mortality. J. Antimicrob. Chemother., 70:2362–2368.

Colombo

A.L.

, Guimarães

, Sukienik

, Pasqualotto

A.C.

, Andreotti

, Queiroz-Telles

, Nouér

S.A.

, and Nucci

. 2014. Prognostic factors and historical trends in the epidemiology of candidemia in critically ill patients: an analysis of five multicenter studies sequentially conducted over a 9-year period. Intensive Care Med., 40:1489–1498.

Perlin

D.S.

, Shor

, and Zhao

. 2015. Update on antifungal drug resistance. Curr. Clin. Microbiol. Rep., 2:84–95.

Perlin

D.S.

2014. Echinocandin resistance, susceptibility testing and prophylaxis: implications for patient management. Drugs, 74:1573–1585.

Pappas

P.G.

, Kauffman

C.A.

, Andes

D.R.

, Clancy

C.J.

, Marr

K.A.

, Ostrosky-Zeichner

, Reboli

A.C.

, Schuster

M.G.

, Vazquez

J.A.

, Walsh

T.J.

, Zaoutis

T.E.

, and Sobel

J.D.

. 2016. Clinical practice guideline for the management of candidiasis: 2016 Update by the Infectious Diseases Society of America. Clin. Infect. Dis., 62:e1–e50.

Emri

, Majoros

, Tóth

, and Pócsi

. 2013. Echinocandins: production and applications. Appl. Microbiol. Biotechnol., 97:3267–3284.

Földi

, Szilágyi

, Kardos

, Berényi

, Kovács

, and Majoros

. 2012. Effect of 50% human serum on the killing activity of micafungin against eight Candida species using time-kill methodology. Diagn. Microbiol. Infect. Dis., 73:338–342.

10.

Földi

, Kovács

, Gesztelyi

, Kardos

, Berényi

, Juhász

, Szilágyi

, Mózes

, and Majoros

. 2012. Comparison of in vitro and in vivo efficacy of caspofungin against Candida parapsilosis, C. orthopsilosis, C. metapsilosis and C. albicans. Mycopathologia, 74:311–318.

11.

Kovács

, Gesztelyi

, Berényi

, Domán

, Kardos

, Juhász

, and Majoros

. 2014. Killing rates exerted by caspofungin in 50% serum and its correlation with in vivo efficacy in a neutropenic murine model against Candida krusei and C. inconspicua. J. Med. Microbiol., 63:186–194.

12.

Domán

, Kovács

, Perlin

D.S.

, Kardos

, Gesztelyi

, Juhász

, Bozó

, and Majoros

. 2015. Dose escalation studies with caspofungin against Candida glabrata. J. Med. Microbiol., 64:998–1007.

13.

Spreghini

, Orlando

, Tavanti

, Senesi

, Giannini

, Manso

, and Barchiesi

. 2012. In vitro and in vivo effects of echinocandins against Candida parapsilosis sensu stricto, Candida orthopsilosis and Candida metapsilosis. J. Antimicrob. Chemother., 67:2195–2202.

14.

Szabó

, Tóth

, Kovács

, Kardos

, Maráz

, Rozgonyi

, and Majoros

. 2008. Evaluation of the new Micronaut-Candida system compared to the API ID32C method for yeast identification. J. Clin. Microbiol., 46:1824–1825.

15.

Clinical and Laboratory Standards Institute. 2008. Reference Method for Broth Dilution Antifungal Susceptibility Testing of Yeasts. Approved Standard. 3rd ed. M27-A3. Clinical and Laboratory Standards Institute, Wayne, PA.

16.

Cantón

, Pemán

, Hervás

, and Espinel-Ingroff

. 2013. Examination of the in vitro fungicidal activity of echinocandins against Candida lusitaniae by time-killing methods. J. Antimicrob. Chemother., 68:864–868.

17.

Cantón

, Espinel-Ingroff

, Pemán

, and del Castillo

. 2010. In vitro fungicidal activities of echinocandins against Candida metapsilosis, C. orthopsilosis, and C. parapsilosis evaluated by time-kill studies. Antimicrob. Agents Chemother., 54:2194–2197.

18.

Pfaller

M.A.

, Diekema

D.J.

, Andes

, Arendrup

M.C.

, Brown

S.D.

, Lockhart

S.R.

, Motyl

, Perlin

D.S.

; CLSI Subcommittee for Antifungal Testing. 2011. Clinical breakpoints for the echinocandins and Candida revisited: integration of molecular, clinical, and microbiological data to arrive at species-specific interpretive criteria. Drug Resist. Updat., 14:164–176.

19.

Pfaller

M.A.

, Boyken

, Hollis

R.J.

, Kroeger

, Messer

S.A.

, Tendolkar

, Jones

R.N.

, Turnidge

, and Diekema

D.J.

. 2010. Wild-type MIC distributions and epidemiological cutoff values for the echinocandins and Candida spp. J. Clin. Microbiol., 48:52–56.

20.

Dufresne

S.F.

, Marr

K.A.

, Sydnor

, Staab

J.F.

, Karp

J.E.

, Lu

, Zhang

S.X.

, Lavallée

, Perl

T.M.

, and Neofytos

. 2014. Epidemiology of Candida kefyr in patients with hematologic malignancies. J. Clin. Microbiol., 52:1830–1837.

21.

Nasar

, Ryan

, Frei

C.R.

, Cota

J.M.

, and Wiederhold

N.P.

. 2013. Influence of serum and albumin on echinocandin in vitro potency and pharmacodynamics. Curr. Fungal Infect. Rep., 7:89–95.

22.

Maki

, Matsumoto

, Watabe

, Iguchi

, Tomishima

, Ohki

, Yamada

, Ikeda

, Tawara

, and Mutoh

. 2008. Use of a serum-based antifungal susceptibility assay to predict the in vivo efficacy of novel echinocandin compounds. Microbiol. Immunol., 52:383–391.

23.

Doman

, Kovacs

, Kardos

, Gesztelyi

, Juhasz

, Bozo

, Kardos

, Saleh

, and Majoros

. 2016. Killing rates of caspofungin in 50 percent serum correlate with caspofungin efficacy against Candida albicans in a neutropenic murine model. Curr. Drug Deliv., 13:255–264.

24.

Goto

, Hara

, Tsurumi

H.K.

, Ogawa

, Kitagawa

, Kanemura

, Kasahara

, Yamada

, Shimizu

, Nakamura

, Matsuura

, and Moriwaki

. 2010. Efficacy and safety of micafungin for treating febrile neutropenia in hematological malignancies. Am. J. Hematol., 85:872–876.

25.

Barchiesi

, Spreghini

, Tomassetti

, Della Vittoria

, Arzeni

, Manso

, and Scalise

. 2006. Effects of caspofungin against Candida guilliermondii and Candida parapsilosis. Antimicrob. Agents Chemother., 50:2719–2727.

26.

Sandoval-Denis

, Pastor

F.J.

, Capilla

, Sutton

D.A.

, Fothergill

A.W.

, and Guarro

. 2014. In vitro pharmacodynamics and in vivo efficacy of fluconazole, amphotericin B and caspofungin in a murine infection by Candida lusitaniae. Int. J. Antimicrob. Agents, 43:161–164.

27.

Sanchis

, Martin-Vicente

, Capilla

, and Guarro

. 2016. Antifungal therapies in murine infections by Candida kefyr. Mycoses, 59:253–258.