In vitro activity of new antimicrobial agents against glycopeptide-resistant Enterococcus faecium clinical isolates from France between 2006 and 2008

Médecine et maladies infectieuses 41 (2011) 405–409

General review

In vitro activity of new antimicrobial agents against glycopeptide-resistant Enterococcus faecium clinical isolates from France between 2006 and 2008 Activité in vitro des nouveaux antibiotiques sur les souches d’Enterococcus faecium résistants aux glycopeptides isolées en France entre 2006 et 2008 R. Bérenger , N. Bourdon , M. Auzou , R. Leclercq , V. Cattoir ∗ Centre national de référence de la résistance aux antibiotiques, laboratoire associé entérocoques, service de microbiologie, CHU Côte-de-Nacre, avenue de la Côte-de-Nacre, 14033 Caen cedex 09, France Received 20 November 2010; received in revised form 3 December 2010; accepted 13 December 2010 Available online 7 May 2011

Abstract Objective. – The aim of this study was to determine the in vitro activity of six new antimicrobial agents against glycopeptide-resistant enterococci (GRE) strains from France. Methods. – Sixty epidemiologically unrelated clinical isolates of Enterococcus faecium (vanA or vanB), received at the National Reference Centre for Enterococci (CNR-Enc) between 2006 and 2008, were studied. The MICs of the following antibiotics were determined by broth microdilution according to Antibiogram Committee of the French Society for Microbiology (CA-SFM) guidelines: quinupristin-dalfopristin (QD), linezolid (LZD), daptomycin (DPT), tigecycline (TGC), ceftobiprole (CFT), and telavancin (TLV). Strains were classified using clinical breakpoints recommended by the CA-SFM (Q-D, LZD, TGC), or the Clinical and Laboratory Standards Institute (DPT). Results. – All strains were susceptible to LZD and DPT (MIC90 , 4 and 2 ␮g/ml, respectively) and only a single strain presented intermediate susceptibility to tigecycline (MIC90 , 0.25 ␮g/ml). Thirty percent of strains were resistant to Q-D (MIC90 , 4 ␮g/ml), and CFT was constantly inactive (MIC90 , 64 ␮g/ml). Finally, TLV showed low-level MICs (MIC90 , 0.5 ␮g/ml) against vanB-positive isolates but not against vanA-positive isolates (MIC90 , 8 ␮g/ml). Conclusion. – Although several antibiotics are still active against GRE, it is essential to maintain an active antimicrobial resistance surveillance for these microorganisms considered as a model of multidrug resistance with a potential to transfer resistance to other bacterial species (e.g. Staphylococcus aureus). © 2011 Elsevier Masson SAS. All rights reserved. Keywords: E. faecium; GRE; VRE; Susceptibility; Resistance

Résumé Objectif. – Le but de cette étude a été de déterminer l’activité in vitro de six nouveaux antibiotiques vis-à-vis de souches d’entérocoques résistants aux glycopeptides (ERG) isolées en France. Méthodes. – Soixante souches cliniques non reliées épidémiologiquement d’Enterococcus faecium (vanA ou vanB), rec¸ues au Centre national de référence pour les entérocoques entre 2006 et 2008, ont été étudiées. Les CMI de l’association quinupristine–dalfopristine (Q-D), du linézolide (LZD), de la daptomycine (DPT), de la tigécycline (TGC), du ceftobiprole (CFT) et de la télavancine (TLV) ont été déterminées par la méthode de microdilution en milieu liquide selon les recommandations du CA-SFM. L’interprétation a été réalisée selon les concentrations critiques proposées par le CA-SFM (Q-D, LZD et TGC) ou le Clinical and Laboratory Standards Institute (DPT). Résultats. – Toutes les souches étaient sensibles au LZD et à la DPT (CMI90 , respectivement à 4 et 2 ␮g/ml) et une seule souche était catégorisée intermédiaire à la TGC (CMI90 à 0,25 ␮g/ml). Trente pour cent des souches étaient résistantes à l’association Q-D (CMI90 à 4 ␮g/ml) et le CFT était constamment inactif (CMI90 à 64 ␮g/ml). Enfin, la TLV présentait des CMI basses sur les souches vanB (CMI90 à 0,5 ␮g/ml) mais pas sur les souches vanA (CMI90 à 8 ␮g/ml).

∗

Corresponding author. E-mail address: [email protected] (V. Cattoir).

0399-077X/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.medmal.2010.12.013

406

R. Bérenger et al. / Médecine et maladies infectieuses 41 (2011) 405–409

Conclusion. – Si plusieurs molécules restent encore actives sur les ERG, il est primordial de surveiller l’activité des antibiotiques sur ces microorganismes qui représentent un modèle de multi-résistance et dont le risque est le transfert de celle-ci à d’autres bactéries pathogènes, comme Staphylococcus aureus. © 2011 Elsevier Masson SAS. Tous droits réservés. Mots clés : E. faecium ; ERG ; ERV ; Sensibilité ; Résistance

1. Introduction

2. Material and methods

Enterococci are commensal bacteria of the gastrointestinal tract of mammals including humans. Even if they are considered as mildly virulent, they are responsible for numerous human infections, especially opportunistic (urinary tract infections, bacteremia, endocarditis, and intra-abdominal infections) [1]. In France, enterococci, the most frequent species of which are Enterococcus faecalis and Enterococcus faecium, accounted for 6.4% of bacteria isolated during nosocomial infections in 2006 [1,2]. For several years, these species have become more and more resistant to antibiotics with the emergence of glycopeptideresistant enterococci (GRE) [3,4]. Resistance to these last resort antibiotics for the treatment of enterococcal infections is often due to the acquisition of vanA or vanB genes [5]. Furthermore, these determinants of resistance are mostly detected in E. faecium which is the main reservoir of resistance in humans [4]. In France, the epidemiological situation has evolved since 2004 with a significant increase of GRE cases notified by the French National Institute for Health Surveillance (InVS) and that of the number of strains received at the National Reference Center for Enterococci (CNR-Enc) [6]. Besides ␤-lactams and glycopeptides, new molecules, whether recently marketed or in development, are active in vitro against GRE [7]. Several protein synthesis inhibitor antibiotics are already on the market, such as the quinupristindalfopristin combination (Synercid® ), first parenteral streptogramin IV; linezolid (Zyvoxid® ), first molecule of the oxazolidinone class; or tigecycline (Tygacil® ), first of the glycylcycline sub-class. Daptomycin (Cubicine® ) belonging to the cyclic lipopeptide family is also available; it acts by disrupting the cytoplasmic membrane of Gram-positive bacteria. Finally, two antibiotics targeting Gram-positive bacteria are under development: ceftobiprole, a novel broad-spectrum cephalosporin with anti-methicillin-resistant Staphylococcus aureus (MRSA) activity, and telavancin a vancomycin derivative belonging to a new antibiotic family, lipoglycopeptides. Even if several of these antibiotics have been available for a few years, little data is available on the susceptibility of French GRE clinical isolates to these molecules, as potential alternative antibiotics. The aim of this study was to determine the in vitro activity of these six novel antibiotics against a representative sample of GRE clinical isolates received at the CNR-Enc between 2006 and 2008.

2.1. Bacterial strains Among the GRE strains received at the CNR-Enc between 2006 and 2008 (n = 917), 60 E. faecium non-epidemiologically related strains (different pulsed-field electrophoresis profiles) were selected: 15 in 2006, 13 in 2007, and 32 in 2008. All strains were isolated in metropolitan France (except for one strain isolated in New-Caledonia). Almost half of these (48.3%) corresponded to colonization or fecal carriage. Finally, all strains were accurately identified at the species-level by multiplex PCR, as previously described [8]. The reference strains used for MIC determination were E. faecalis ATCC 29212 and S. aureus ATCC 29213. 2.2. Disk diffusion susceptibility testing Antibiograms were performed on Mueller-Hinton (MH) medium according to Antibiogram Committee of the French Society for Microbiology (CA-SFM) recommendations as well as the interpretation of results [9]. The 11 tested antibiotics were the following (Bio-Rad disks): ampicillin (10 ␮g), kanamycin (1 mg), gentamicin (500 ␮g), chloramphenicol (30 ␮g), doxycycline (30 UI), erythromycin (15 UI), lincomycin (15 ␮g), pristinamycin (15 ␮g), sulfamethoxazoletrimethoprim (1.25/3.75 ␮g) or cotrimoxazole, levofloxacin (5 ␮g), and rifampicin (30 ␮g). 2.3. MIC determination The MICs of vancomycin and teicoplanin were determined by the E-test method (AB Biodisk, Solna, Suede) according to the manufacturer’s recommendations. The MICs of the six following antibiotics were determined by the broth microdilution method according to CA-SFM recommendations [10]: quinupristin-dalfopristin (Q-D), linezolid (LZD), daptomycin (DPT), tigecycline (TGC), ceftobiprole (CFT), and telavancin (TLV). The antibiotic powders were obtained from each pharmaceutical manufacturer. The MICs tested ranged from 0.06 to 64 ␮g/ml. For DPT, the final concentration in Ca2+ ions was adjusted to 50 ␮g/ml. The interpretation of results was made by using breakpoints determined by the CA-SFM for Q-D (susceptible if MIC ≤ 1 ␮g/ml, resistant if MIC > 4 ␮g/ml), LZD (susceptible if MIC ≤ 4 ␮g/ml, resistant if MIC > 4 ␮g/ml), and TGC (susceptible if MIC ≤ 0.25 ␮g/ml, resistant if MIC > 0.5 ␮g/ml) [9], or by the Clinical and Laboratory Standards Institute (CLSI) for DPT (susceptible if

R. Bérenger et al. / Médecine et maladies infectieuses 41 (2011) 405–409

MIC ≤ 4 ␮g/ml). No breakpoint is currently available for CFT and TLV [11]. 2.4. Detection of van genes The various genes of the van alphabet (vanA, vanB, vanC, vanD, vanE, and vanG) were screened for by multiplex PCR, as previously described [8]. 3. Results Among the 60 E. faecium strains studied, 44 (73.3%) carried the vanA gene, 15 (25%) the vanB gene, and one (1.7%) the vanD gene. No vanC, vanE, or vanG genes were detected. All vanB strains were resistant to vancomycin (MIC50 and MIC90 , 16 and 48 ␮g/ml, respectively) and all vanA strains were resistant both to vancomycin (MIC50 and MIC90 > 256 ␮g/ml) and to teicoplanin (MIC50 and MIC90 , 48 and 256 ␮g/ml, respectively). Most strains were resistant to ␤-lactams, kanamycin, fluoroquinolones, macrolides-lincosamides and more than half were resistant to gentamicin, doxycycline, cotrimoxazole, and rifampicin (Table 1). Finally, nearly 80 and 95% of the strains

407

Table 1 Proportion of E. faecium clinical isolates with intermediate susceptibility or resistant to various antimicrobial agents tested by the disk diffusion method. Proportions de souches cliniques d’E. faecium résistants ou intermédiaires aux différents antibiotiques testées par diffusion en milieu gélosé. Antibiotica

All strains (n = 60)

vanA strains (n = 44)

vanB strains (n = 15)

Pb

Ampicillin Kanamycin Gentamicin Chloramphenicol Doxycycline Erythromycin Lincomycin Pristinamycin Cotrimoxazole Levofloxacin Rifampicin

93.3 88.3 60 21.7 61.7 98.3 90 6.7 61.7 91.7 55

90.9 86.4 59.1 25 70.5 100 95.5 6.8 56.8 90.9 54.5

100 93.3 66.7 13.3 33.3 93.3 73.3 6.7 80 93.3 60

0.56 0.67 0.76 0.48 0.01 0.25 0.03 1.0 0.13 1.0 0.77

a The interpretation of results was performed using critical diameters determined by the CA-SFM. b P value calculated with Fisher’s exact test, bilaterally between groups of vanA and vanB strains.

Table 2 In vitro activity of the six new antimicrobial agents against the 60 E. faecium clinical isolates. Activité in vitro des six nouveaux antibiotiques sur les 60 souches cliniques d’E. faecium. Bacteria (number of clinical isolates) and antibiotics

% of susceptiblea strains

MIC (␮g/ml) MIC50

MIC90

Interval

All strains (n = 60) Quinupristin-dalfopristin Linezolid Daptomycin Tigecycline Ceftobiprole Telavancin

0.5 2 1 0.25 64 4

4 4 2 0.25 64 8

0.06–16 0.5–4 0.25–4 0.12–0.5 1– > 64 0.06–16

70 100 100 98.3 – –

vanA strains (n = 44) Quinupristin-dalfopristin Linezolid Daptomycin Tigecycline Ceftobiprole Telavancin

0.5 2 1 0.25 32 4

4 4 2 0.25 64 8

0.06–16 0.5–4 0.25–4 0.12–0.5 1– > 64 0.25–16

79.5 100 100 97.7 – –

vanB strains (n = 15) Quinupristin-dalfopristin Linezolid Daptomycin Tigecycline Ceftobiprole Telavancin

0.5 2 1 0.25 64 0.06

4 4 2 0.25 64 0.5

0.12–4 0.5–4 0.5–2 0.12–0.25 32–64 0.06–4

60 100 100 100 – –

2 2 1 0.125 64 1

– – – – – –

vanD strain (n = 1) Quinupristin-dalfopristin Linezolid Daptomycin Tigecycline Ceftobiprole Telavancin

– – – – – –

– – – – – –

a The interpretation of results was performed using breakpoints determined by the CA-SFM (for quinupristin–dalfopristin, linezolid, and tigecycline) or the CLSI (for daptomycin). No breakpoint has been validated yet for ceftobiprole and telavancin.

408

R. Bérenger et al. / Médecine et maladies infectieuses 41 (2011) 405–409

were susceptible to phenicols and streptogramins, respectively (Table 1). The rate of resistance was not significantly different for most antibiotics tested, except for doxycycline and lincomycin for which the proportions of resistant vanA-positive strains were higher (Table 1). All the strains were susceptible to LZD (MIC50 and MIC90 , 2 and 4 ␮g/ml, respectively) and to DPT (MIC50 and MIC90 , 1 and 2 ␮g/ml, respectively) (Table 2). Only one strain was of intermediate susceptibility to TGC with a MIC at 0.5 ␮g/ml (MIC50 and MIC90 , 0.25 ␮g/ml) (Table 2). However, 28.3% (n = 17) and 1.7% (n = 1) of the strains were respectively of intermediate susceptibility and resistant to the Q-D combination (MIC50 and MIC90 , 0.5 and 4 ␮g/ml, respectively) (Table 2). Among these 17 strains of intermediate susceptibility, nine had a MIC at 2 ␮g/ml and eight at 4 ␮g/ml while the resistant strain presented a high MIC at 16 ␮g/ml. For CFT, most strains had high MICs (MIC50 and MIC90 , 64 ␮g/ml) (Table 2) even if some of them had much lower MICs, at 1 or 2 ␮g/ml. MICs of TLV were rather high (MIC50 and MIC90 , 4 and 8 ␮g/ml, respectively) even if they were distributed over a wide range from 0.06 to 16 ␮g/ml. It should be noted that there was no significant difference between MIC50 and MIC90 values of the vanA strain group and that of vanB strains, except for TLV. Indeed, this molecule had low MICs against most vanB strains (MIC50 and MIC90 , 0.06 and 0.5 ␮g/ml, respectively) whereas most vanA strains seemed much less susceptible (MIC50 and MIC90 , 4 and 8 ␮g/ml, respectively) (Table 2). 4. Discussion One of the CNR-Enc missions is to monitor resistance to antibiotics of GRE strains (epidemic or not) isolated in France. By selecting a representative panel of strains isolated between 2006 and 2008, we determined the in vitro activity of antibiotics recently marketed or in development which are therapeutic alternatives for the management of infections due to these bacteria. It should be noted that the predominance of vanA strains in our sample is representative of all strains received at the CNR-Enc between 2006 and 2008 even if the prevalence of vanB strains was higher in 2008 [6]. According to disk-diffusion antibiograms, most GRE strains presented co-resistance to other antibiotic families, especially ␤lactams, macrolides, and fluoroquinolones. This was expected since almost all GRE strains responsible for outbreaks throughout the world belong to clonal complex 17 (CC17), which exhibits high-level resistance to ␤-lactams and fluoroquinolones [3,4,12]. Furthermore, resistance to erythromycin, most often mediated by the ermB gene, is also very frequent among GRE [13]. All strains were susceptible to LZD, as expected since resistance to this antibiotic is rare in Enterococcus spp. and has never been reported in France. Only a few strains have been reported as resistant, particularly in the US [14]. Resistance to LZD is mostly due to ribosomal mutations (frequency of in vitro selection < 10−9 ) but also possibly due to the acquisition of the plasmid-mediated cfr gene, coding for a ribosomal RNA methylase conferring cross resistance to five antibi-

otic families (PhLOPSA phenotype): phenicols, lincosamides, oxazolidinones, pleuromutilins, and streptogramins A [15]. It should be noted that the cfr gene was screened for by PCR on the 60 strains in our study but none carried it (results not shown). All strains were also susceptible to DPT with MIC50 and MIC90 similar to those previously reported [16,17]. It should be noted that resistance to this antibiotic is rare in enterococci and only a few resistant clinical isolates, with MICs superior or equal to 32 ␮g/ml, have been reported [18]. It seems that the E. faecium resistance mechanism to DPT differs from that of S. aureus [19,20]. TGC also seems to be active on GRE strains with low MICs (MIC90 , 0.25 ␮g/ml) similar to those previously described (MIC90 , 0.125 ␮g/ml) [17]. It should be noted that if 60% of tested strains are resistant to older tetracyclines, TGC remains active against these strains resistant to doxycycline (by efflux and/or ribosomal protection) [7]. Finally, resistance to TGC seems to be very difficult to select in vitro (frequency < 10−9 ) and only one E. faecalis clinical isolate (with a MIC at 1 ␮g/ml by broth microdilution) has been reported [21]. The Q-D combination seems the least interesting alternative treatment for GRE infections. Indeed, around 30% of tested E. faecium strains (an intrinsically-susceptible species) were not susceptible to the Q-D combination, a much higher rate than the 10% reported in Europe [22]. These strains could express an lincosamide-streptogramin A (LSA ) phenotype. Nevertheless, resistance is still not explained in E. faecium whereas in E. faecalis (an intrinsically-resistant species), production of an ABC protein coded by the lsa gene is responsible for this LSA phenotype [23]. As expected, CFT is not active against most enterococcal strains tested. Indeed, this molecule, even if it is significantly active on MRSA, belongs to the cephalosporin family and enterococci are intrinsically resistant to this group of ␤-lactams [7]. Nevertheless, several strains (n = 7) surprisingly exhibit low MICs at 1 or 2 ␮g/ml. The in vitro activity of TLV depends on the type of resistance for the tested strains, with low MICs against vanB strains and higher MICs against vanA strains. This correlates to published data indicating that telavancin seems active against vanB or vanC strains but not against vanA strains [7,24]. 5. Conclusion Enterococci are responsible for many opportunistic infections and their multidrug resistance is a real healthcare issue. Since 2004, the number of notified cases of GRE has dramatically increased in France and numerous outbreaks have been reported. Even if several molecules remain active against GRE, the epidemiological data of infections due to these microorganisms should be strictly monitored, especially given the risk of transferring multidrug resistance to highly pathogenic bacterial strains such as S. aureus. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.

R. Bérenger et al. / Médecine et maladies infectieuses 41 (2011) 405–409

References [1] Murray BE. The life and times of the Enterococcus. Clin Microbiol Rev 1990;3:46–65. [2] Enquête nationale de prevalence des infections nosocomiales, France, juin 2006. (www.invs.sante.fr/enp2006). [3] Willems RJ, Bonten MJ. Glycopeptide-resistant enterococci: deciphering virulence, resistance and epidemicity. Curr Opin Infect Dis 2007;20:384–90. [4] Werner G, Coque TM, Hammerum AM, Hope R, Hryniewicz W, Johnson A, et al. Emergence and spread of vancomycin resistance among enterococci in Europe. Euro Surveill 2008;13(47), PII: 19046. [5] Courvalin P. Vancomycin resistance in gram-positive cocci. Clin Infect Dis 2006;42(Suppl 1):S25–34. [6] Bourdon N, Fines-Guyon M, Auzou M, Leclercq R, Cattoir V. Epidemiological and molecular characterization of vancomycinresistant enterococci (VRE) from France collected between 2006 and 2008, abstr. C2-1899. 49th Intersc. Conf Antimicrob Agents Chemother 2009. [7] Cattoir V, Daurel C. Quelles nouveautés en antibiotherapie ? Med Mal Infect 2010;40:135–54. [8] Depardieu F, Perichon B, Courvalin P. Detection of the van alphabet and identification of enterococci and staphylococci at the species level by multiplex PCR. J Clin Microbiol 2004;42:5857–60. [9] Comite de l’antibiogramme de la Société franc¸aise de microbiologie (CASFM), recommandations 2010 (www.sfm.asso.fr). [10] Comité de l’antibiogramme de la Société franc¸aise de microbiologie (CASFM). Technical recommendations for in vitro susceptibility testing. Clin Microbiol Infect 1996;2(Suppl. 1):S11–25. [11] Clinical, Laboratory Standards, Institute. Performance standards for antimicrobial susceptibility testing - Eighteenth informational supplement: M100-S18. USA: CLSI, Wayne, PA; 2008. [12] Top J, Willems R, Bonten M. Emergence of CC17 Enterococcus faecium: from commensal to hospital-adapted pathogen. FEMS Immunol Med Microbiol 2008;52:297–308. [13] Cetinkaya Y, Falk P, Mayhall CG. Vancomycin-resistant enterococci. Clin Microbiol Rev 2000;13:686–707.

409

[14] Perry CM, Jarvis B. Linezolid: a review of its use in the management of serious gram-positive infections. Drugs 2001;61:525–51. [15] Long KS, Poehlsgaard J, Kehrenberg C, Schwarz S, Vester B. The Cfr rRNA methyltransferase confers resistance to Phenicols, Lincosamides, Oxazolidinones. Pleuromutilins, and Streptogramin A antibiotics. Antimicrob Agents Chemother 2006;50:2500–5. [16] Richter SS, Kealey DE, Murray CT, Heilmann KP, Coffman SL, Doern GV. The in vitro activity of daptomycin against Staphylococcus aureus and Enterococcus species. J Antimicrob Chemother 2003;52:123–7. [17] Rathe M, Kristensen L, Ellermann-Eriksen S, Thomsen MK, Schumacher H. Vancomycin-resistant Enterococcus spp.: validation of susceptibility testing and in vitro activity of vancomycin, linezolid, tigecycline and daptomycin. APMIS 2010;118:66–73. [18] Lewis 2nd JS, Owens A, Cadena J, Sabol K, Patterson JE, Jorgensen JH. Emergence of daptomycin resistance in Enterococcus faecium during daptomycin therapy. Antimicrob Agents Chemother 2005;49:1664–5. [19] Boucher HW, Sakoulas G. Perspectives on Daptomycin resistance, with emphasis on resistance in Staphylococcus aureus. Clin Infect Dis 2007;45:601–8. [20] Montero CI, Stock F, Murray PR. Mechanisms of resistance to daptomycin in Enterococcus faecium. Antimicrob Agents Chemother 2008;52:1167–70. [21] Werner G, Gfrörer S, Fleige C, Witte W, Klare I. Tigecycline-resistant Enterococcus faecalis strain isolated from a German intensive care unit patient. J Antimicrob Chemother 2008;61:1182–3. [22] Deshpande LM, Fritsche TR, Moet GJ, Biedenbach DJ, Jones RN. Antimicrobial resistance and molecular epidemiology of vancomycin-resistant enterococci from North America and Europe: a report from the SENTRY antimicrobial surveillance program. Diagn Microbiol Infect Dis 2007;58:163–70. [23] Singh KV, Weinstock GM, Murray BE. An Enterococcus faecalis ABC homologue (Lsa) is required for the resistance of this species to clindamycin and quinupristin-dalfopristin. Antimicrob Agents Chemother 2002;46:1845–50. [24] Krause KM, Renelli M, Difuntorum S, Wu TX, Debabov DV, Benton BM. In vitro activity of telavancin against resistant gram-positive bacteria. Antimicrob Agents Chemother 2008;52:2647–52.