AAC Accepted Manuscript Posted Online 28 December 2015 Antimicrob. Agents Chemother. doi:10.1128/AAC.01763-15 Copyright © 2015, American Society for Microbiology. All Rights Reserved. 1 Post treatment antifungal resistance among colonizing Candida isolates in candidemia patients: 2 results from a systematic multicenter study 3 Jensen R. H.1, Johansen H. K.2, Søes L. M.3,4, Lemming L.E.5, Rosenvinge F. S.6,7, Nielsen L.8, Olesen B.9, 4 Kristensen L.10,11, Dzajic, E.12, Astvad K. M. T.1 & *Arendrup M. C.1 D o w n 5 1Unit of Mycology, Department of Microbiology and Infection Control, Statens Serum Institut, Copenhagen, lo a d 6 Denmark and Departments of Clinical Microbiology at e d f r 7 2Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark, 3Hvidovre University Hospital, o m 8 Hvidovre, Denmark, 4Slagelse Sygehus, Slagelse, Denmark, 5Aarhus University Hospital, Aarhus, Denmark, h t t 9 6Sygehus Lillebælt, Vejle, Denmark, 7Odense University Hospital, Odense, Denmark, 8Herlev University p : / 10 Hospital, Herlev, Denmark, 9Nordsjællands Hospital, Hillerød, Denmark, 10Hospitalsenheden Vest, Herning, /a a 11 Denmark, 11Hospitalsenheden Midt, Viborg, Denmark, 12Sydvestjysk Sygehus, Esbjerg, Denmark c. a s m 12 . o r g / 13 *Corresponding author: Maiken Cavling Arendrup o n 14 Microbiology and Infection Control J a n 15 Statens Serum Institut u a r y 16 Artillerivej 5 2 9 17 Building 43-317 , 2 0 18 2300 Copenhagen 1 9 b 19 Denmark y g 20 Email: [email protected] Tel: (+45) 32 68 32 23 u e s 21 t 22 Key words: antifungal resistance, Candida colonization, Candida glabrata, genotyping 23 Abstract 24 The prevalence of intrinsic and acquired resistance among colonizing Candida isolates from patients after 25 candidemia was systematically investigated in a 1-year nationwide study. Patients were treated at the 26 discretion of the treating physician. Oral swabs were obtained after treatment. Species distributions and D 27 MIC data were investigated for blood and post-treatment oral isolates from patients exposed to <7 or ≥7 o w n 28 days of either azoles or echinocandins. Species identification was confirmed using MALDI-TOF MS and ITS lo a d 29 sequencing, susceptibility was examined by EUCAST E.Def 7.2 methodology, echinocandin resistance by FKS e d 30 sequencing and genetic relatedness by MLST analyses. Onehundrednintythree episodes provided 205 blood fr o m 31 and 220 oral isolates. MLST analysis demonstrated genetic relationship in 90% of all paired blood and oral h t t p 32 isolates. Patients exposed to ≥7 days of azoles (N=93) had a significantly higher proportion of species : / / a 33 intrinsically less susceptible to azoles (particularly Candida glabrata) among oral isolates compared to initial a c . a 34 blood isolates (36.6% vs 12.9%, P<0.001). A similar shift towards species less susceptible to echinocandins s m . 35 among 85 patients exposed to ≥7 days of echinocandins was not observed (oral: 4.8% vs Blood: 3.2%, o r g / 36 P>0.5). Acquired resistance in Candida albicans was rare (<5%). However, acquired resistance to o n J 37 fluconazole (29.4%, P<0.05) and anidulafungin (21.6%, P<0.05) was common in C. glabrata from patients a n u 38 exposed to either azoles or echinocandins. Our findings suggest the colonizing mucosal flora may be an a r y 39 unrecognized reservoir of resistant Candida species, especially C. glabrata following treatment for 2 9 , 40 candidemia. The resistance rates were high raising concern in general for patients exposed to antifungal 2 0 1 41 drugs. 9 b y g u e s t 42 Introduction 43 Candidemia due to Candida species, which are intrinsically less susceptible to fluconazole are becoming 44 proportionally more common at the expense of Candida albicans (1–5). Prophylactic antifungal use has 45 been demonstrated to significantly drive the species distributions in candidemia patients towards less D 46 susceptible species such as Candida glabrata in patients exposed to fluconazole and Candida parapsilosis o w n 47 when exposed to caspofungin (6–8). Indeed, intrinsic resistance and the changing epidemiology is the main lo a d 48 concern. On the other hand, in epidemiologic candidemia studies acquired resistance is rare, perhaps in e d 49 part because such studies only include the initial blood isolate from the candidemia patient. Thus, fr o m 50 subsequent and potentially resistant isolates may be overlooked. Resistance selection may be facilitated by h t t p 51 longer duration of treatment (9) and sub-therapeutic drug levels (10, 11). Hence, abdominal candidiasis and : / / a 52 Candida species at mucosal surfaces may represent a nest and reservoir for resistant isolates (10, 12). a c . a 53 Typing studies have illustrated that the concomitant and predominant colonizing Candida in candidemia s m . 54 patients is also the invasive Candida species (90-95% concordance) and most patients are still colonized o r g / 55 despite that the infection is cleared (7, 9, 13–17). Thus, the persistent colonizing flora is a relevant reservoir o n J 56 for resistant Candida and may impose a risk in this group of patients concerning reinfection but also in a n u 57 other patient settings where infection may follow prior antifungal drug exposure. This study is a systematic a r y 58 nationwide multicenter study investigating the colonizing flora obtained after antifungal drug exposure in 2 9 , 59 candidemia patients. The purpose was to elucidate the prevalence of Candida species less susceptible to 2 0 1 60 antifungals and the extent of acquired resistance in the colonizing mucosal flora of patients treated for 9 b y 61 candidemia, since the increasing number of reports addressing these topics are mainly case reports or g u e 62 single center studies (18). s t 63 Materials and methods 64 Patients, isolates and treatment. All candidemia patients in Denmark diagnosed within the 1-year study 65 period (March 2013-February 2014) were eligible for inclusion in this study. Blood isolates were referred to 66 the Mycology Reference Laboratory at Statens Serum Institut as part of the ongoing national surveillance of D 67 fungaemia from all departments of clinical microbiology in Denmark. Corresponding candidemia patients o w n 68 were located through the Danish Microbiology Database (MiBa) and one oral swab was requested after lo a d 69 discontinuation of antifungal treatment. Separate candidemia episodes were only included if occurring >21 e d 70 days apart. Information on antifungal treatment for invasive candidiasis (IC) but not empirical treatment fr o m 71 was sought. Patients were excluded if an oral swab or IC treatment information was not obtained. h t t p : 72 Prevalence of intrinsic resistance: In patients (N=93) treated with ≥7 days of azoles but <7 days of another // a a 73 drug class (Figure 1), species distributions in blood isolates (before IC treatment) and oral isolates (after IC c . a s 74 treatment) were determined and the proportions of species intrinsically less susceptible to azoles were m . o r 75 compared. Likewise, in patients (N=62) treated with ≥7 days of echinocandins but <7 days of another drug g / o 76 class (Figure 1), the proportions of species intrinsically less susceptible to echinocandins among blood and n J a 77 oral isolates were compared. The 7-day cut-off for significant exposure was selected based on previous n u a 78 findings identifying this duration a critical driver of altered species distribution (9). As co-administration of r y 2 79 different drug classes (combination or sequential) for ≥7 days might bias the evaluation of each drug class’ 9 , 2 80 impact on post treatment species distribution those patients (N=26) were excluded from this evaluation. 0 1 9 b 81 Species specific prevalence of acquired resistance: Rates of acquired resistance among initial blood and y g u 82 post treatment oral isolates were evaluated for C. albicans and C. glabrata isolates. Susceptibility data of e s t 83 oral isolates from patients exposed to ≥7 days of either azoles (N=114, 109/114 fluconazole and 5/114 84 voriconazole) or echinocandins (N=85) was evaluated for both species. Susceptibility data from blood 85 isolates as well as oral isolates exposed to <7 days of either azoles (N=79) or echinocandins (N=108), 86 respectively, was included for both species for comparison. Co-administration of unrelated drug classes was 87 not regarded a driver of acquired resistance. Therefore the evaluation of acquired resistance to a given 88 drug class included all patients with and without exposure to alternative drug classes. 89 Ethical considerations. The study was approved by the Research Ethics Committees of the Capital Region of 90 Denmark (protocol no. H-4-2012-FSP 104), the Danish Data Protection Agency (J.no. 2013-41-1606) and the D 91 Danish Health and Medicines Authority (J. no. 3-3013-302/1/). o w n lo 92 Microbiology. Blood isolates and oral swabs were inoculated on CHROMagar (CHROMagar CO., Paris, a d e 93 France) for 2-4 days at 37oC. Species identifications were done by conventional methods and MALDI-TOF d f r 94 MS analysis (Bruker Daltonics, Germany), supplemented with carbon assimilation with ATB strips (ID32C; o m h 95 bioMérieux, Marcy l’Etoile, France) or internal transcribed spacer (ITS) DNA sequencing when necessary t t p : 96 (19). Polyfungal samples were identified by colony color differentiation (also within the same species) and // a a 97 separated by subculturing. Isolates were regarded unique if a different species was detected or if the same c . a s 98 species had different genotypes or susceptibility profile (≥three 2-fold dilution MIC difference to any m . o r 99 antifungal tested). MiBa was further used to assess microbiological data. g / o n 100 Susceptibility testing was done according to EUCAST E.Def 7.2 (azoles and anidulafungin) (20) and by Etest J a n 101 (amphotericin B). Susceptibility interpretation was performed applying the defined EUCAST breakpoints u a r y 102 (21, 22) and http://www.eucast.org/clinical_breakpoints/. 2 9 , 2 103 Statistical analyses. Species distributions were compared by unpaired proportion analyses (Chi-squared 0 1 9 104 tests). Candida krusei, C. glabrata, Candida guilliermondii, Candida pelliculosa, Saccharomyces cerevisiae b y 105 and Cryptococcus neoformans were classified as species intrinsically less susceptible to azoles. Similarly, g u e 106 C. parapsilosis, C. guilliermondii and S. cerevisiae were classified as less susceptible to the echinocandins. S. s t 107 cerevisiae was regarded less susceptible to both drug-classes because the modal MIC is comparable to C. 108 glabrata for azoles (EUCAST data) but at least one step higher for echinocandins (unpublished data). When 109 multiple different species were found in blood or swabs, the isolate of the species with the highest ECOFF 110 was counted. The number of negative oral cultures or polyfungal oral cultures was compared as a sub- 111 analysis between patients exposed to either azoles or echinocandins to determine the degree of fungal 112 colonization after antifungal exposure.. 113 Fisher´s exact test of unpaired data was applied to compare proportions of acquired resistance among 114 C. albicans and C. glabrata isolates from blood and oral swabs from patients exposed to <7 days or ≥7 days D 115 of antifungal. The difference in mean MICs for each drug-class was also compared for C. glabrata and o w n 116 C. albicans analyzed by Wilcoxon rank-sum test, assuming non-normal distribution of MIC data. All lo a d 117 polyfungal isolates were included for these analyses. e d f r 118 All P-values <0.05 were considered significant. R was used for all statistical calculations. o m h t 119 Multi-locus sequence typing. MLST was performed as previously described for the six major species C. tp : / / 120 albicans (23), C. glabrata (24), C. krusei (25), C. tropicalis (26), C. dubliniensis (27), S. cerevisiae (28) and a a c . 121 microsatellite typing was used for C. parapsilosis (29). MLST profiles were assigned according to a s m 122 corresponding pubMLST databases, when available or assigned as novel for S. cerevisiae starting from ”1” . o r g 123 for each allele. CLC Main Workbench (CLC-Bio, Qiagen) bioinformatics software with the MLST-plugin was / o n 124 applied to analyze sequence data. One allelic difference were considered microevolution between paired J a n 125 isolates based on previous findings and thus isolates were regarded as genetically unrelated when differing u a r y 126 at two or more alleles for all MLST schemes (30, 31). 2 9 , 2 127 Resistance mechanisms. Underlying molecular resistance mechanisms were characterized for echinocandin 0 1 9 128 resistance by sequencing of the gene encoding β-1,3-glucan synthase (FKS1). b y g u e s t 129 Results 130 Demographics. During the 1-year study period, 471 unique blood isolates from 455 candidemia cases were 131 referred to the reference laboratory. Oral swabs were obtained from 200 of these patients from all 12 132 major hospitals in Denmark, 63 (31.5%) were from ICU, 55 (27.5%) from surgical, 73 (36.5%) from medical D 133 and 9 (4.5%) from other units. In total 194 out of 455 patients (42.6%) deceased before the swab was o w n 134 obtained, and thus 76.6% (200 out of 261) of all eligible candidaemia patients were included (Figure 1). The lo a d 135 remaining 61 patients were discharged before a mouth swab was obtained. Neither median age (67 vs 68 e d 136 years) nor proportion of men (59.6% vs 57.7%) varied significantly between enrolled patients and the total fr o m 137 population of candidemia patients. Due to insufficient treatment information, seven patients were h t t p 138 excluded. In total 190 patients (including one neonate) and 193 episodes provided 205 blood isolates and : / / a 139 220 swab isolates included for statistical analyses. Total IC treatment median duration (range) was 17 days a c . a 140 (5-61 days). The duration of the individual drug classes in patients treated with ≥7 days of azoles or s m . 141 echinocandins, separately, was 14 days (7-40 days) for azoles and 15 days (7-56 days) for echinocandins, o r g / 142 respectively. o n J a 143 Species distributions. The species distribution among blood isolates from included patients was similar to n u a 144 the distribution for all candidemia cases (Figure 2). Among 93 patients treated for ≥7 days with azoles but r y 2 145 <7 days of another drug class, 66 (71.0%) had C. albicans in the original blood culture, while among the 62 9 , 2 146 patients treated with ≥7 days of echinocandins and <7 days of another drug class; 41 (66,1%) had C. 0 1 9 147 glabrata in the blood (Figure 2). In azole exposed patients, the species distributions from blood and oral b y g 148 isolates were significantly different with 71.0% C. albicans in blood and less than 20% oral isolates (Figure u e s 149 2). On the other hand, C. glabrata and S. cerevisiae constituted about 10% and 1% respectively in blood but t 150 about 25% and 7.5% respectively in oral isolates. The overall number of isolates less susceptible to 151 fluconazole were significantly higher (P<0.001) in the oral isolates 34/93 (36.6%) compared to the blood 152 isolates 12/93 (19.9%). In the echinocandin group minor differences were observed when comparing 153 species distributions among blood and oral isolates from exposed patients where a lower proportion of C. 154 albicans and C. glabrata was observed. Only 2 (3.2%) isolates categorized as intrinsically less susceptible to 155 echinocandins were found in blood isolates (1 C. parapsilosis and 1 Geotrichum candidum) in contrast to 156 three (4.8%) (1 C. parapsilosis, 1 C. guilliermondii and 1 S. cerevisiae) in oral isolates. D 157 Of patients treated with an echinocandin 87% (54 out of 62) had culture positive oral swabs and 63% o w n 158 (39/62) had a polyclonal colonization, compared to 57% (53/93) and 31% (29/93) respectively among lo a d 159 patients treated with azoles. Both proportions of culture positive oral swabs and polyclonal colonization e d 160 were significantly higher in the echinocandin group (P<0.001 and P<0.001, respectively). fr o m h 161 MLST analyses. Among 193 patients, 104 (53.9%) patients harbored a total of 112 paired isolates, where t t p : 162 the same species was found before and after antifungal exposure, 32 patients (16.5%) had a different // a a 163 colonizing species post treatment, 57 patients (29.5%) were swab culture negative and 72 (37.3%) had two c . a s 164 or more isolates in the oral samples. Only 43/100 (43%) of patients with C. albicans in the blood were m . o r 165 colonized with C. albicans after treatment, two patients with C. albicans in the blood had two different C. g / o 166 albicans in the oral sample and one patient with two separate C. glabrata candidemia episodes had C. n J a 167 albicans in both oral swabs 10 weeks apart (Table S1). Among the total of 45 paired C. albicans isolates n u a 168 MLST analysis showed that 38 (84.4%) had similar genotypes. Similarly, 52 out of 62 (83.9%) patients with r y 2 169 C. glabrata were colonized with C. glabrata after treatment, four C. glabrata patients had two different C. 9 , 2 170 glabrata isolates in the oral sample and among a total of 56 paired C. glabrata isolates 55 (98%) had similar 0 1 9 171 genotypes. Overall, 101/112 (90%) of paired Candida isolates had similar genotypes (Table 1 and Table S1). b y g u 172 Susceptibility and resistance. Fluconazole MIC distributions for the 62 blood and 82 oral C. glabrata isolates e s t 173 were bimodal with peaks at 2-4 mg/L and 32-64 mg/L (Figure 3). The proportion of resistant isolates (MIC > 174 32 mg/L) was significantly higher among oral isolates from azole exposed patients (29.4%) compared to 175 initial blood isolates (4.8%, P<0.01) as well as to oral isolates from less exposed patients (10.4%, P<0.05) 176 (Table 2). Accordingly, the geometric mean MIC for fluconazole was significantly higher for isolates from 177 azole exposed patients (Table 2). A similar trend was observed for voriconazole MIC distributions although 178 no clinical breakpoint has yet been established for C. glabrata (Figure 3). For anidulafungin, the modal MIC 179 was 0.06 mg/L and again the proportion of isolates with MICs above the breakpoint was significantly higher 180 among oral isolates from patients exposed to echinocandins (21.6%) compared to blood and to oral isolates 181 from patients exposed <7 days to echinocandins (4.8% and 3.2%), respectively (Table 2). No statistical D o w 182 differences were observed when comparing geometric means (Table 2). Among the 11 C. glabrata isolates n lo a 183 (21.6%) with anidulafungin MIC above the breakpoint (Table 2), three isolates were FKS wild-type (with d e d 184 anidulafungin MICs of 0.125 mg/L), while eight isolates harbored well-known mutations responsible for the f r o 185 observed resistance (MICs 0.125 - >1 mg/L) (Figure 3). Micafungin MICs were above breakpoint for six oral m h 186 isolates from patients exposed to an echinocandin all of which harbored FKS mutations. For two FKS tt p : / / 187 mutants (F659L and F659I) the micafungin MICs were below the breakpoint (both 0.015 mg/L). Two C. a a c 188 glabrata isolates from blood harbored FKS hot-spot mutations (conferring amino acid substitutions F659- .a s m 189 DEL and S663F+L630Q). . o r g 190 Only 1 out of 47 (2.1%) C. albicans oral isolate from a patient exposed to 16 days of fluconazole was azole / o n 191 resistant, while 0 out of 37 oral isolates from unexposed patients and none out of 100 blood isolates were J a n 192 resistant. Two out of 43 (4.7%) C. albicans isolates from patients exposed to echinocandins harbored an FKS u a r 193 mutation (causing F641L and D648V amino acid substitutions, respectively) and were micafungin resistant, y 2 9 194 one of which with cross-resistance to anidulafungin (Figure 3). Both FKS mutants were paired with a , 2 0 195 previous isogenic wild-type isolate and derived from patients who had received 28 and 15 days of 1 9 b 196 caspofungin respectively before the resistant isolate was obtained (Table S1). MIC distributions of y g u 197 genetically linked blood and oral isolates for both C. albicans and C. glabrata displayed equivalent MIC e s t 198 histograms as for the overall susceptibility data, which also included isolates solely found in either blood or 199 oral swabs (Figure S1). 200 MIC data for other Candida species showed that two patients had persistent C. krusei isolates in blood and 201 oral swabs displaying anidulafungin resistance (MIC: 0,125 mg/L). Both isolates harbored the L701M Fks1 202 substitution (Table 3). No other Candida species displayed acquired resistance to echinocandins. Two oral 203 isolates were fluconazole resistant (1 out of 8 C. dubliniensis) and 1 out of 3 C. parapsilosis while 1 out of 8 204 C. dubliniensis isolate was categorized as intermediate, all of which were from patients exposed to >14 days 205 of fluconazole. In addition, 1 out of 8 C. tropicalis from blood and 2 out of 12 C. tropicalis oral isolates 206 displayed trailing phenotypes (partial growth inhibition) and these three isolates were acquired from D o w 207 patients with no knowledge on previous fluconazole exposure (Table 3). n lo a d 208 Discussion e d f r 209 This study investigated the extent of intrinsic and acquired antifungal resistance of the oral fungal o m h 210 community after antifungal treatment. Overall, this study presented a high proportion of species less t t p : 211 susceptible to azoles in the colonizing flora (primarily C. glabrata) in patients treated with azoles as well as // a a 212 high frequencies of acquired resistance to both azoles and echinocandins among colonizing C. glabrata c . a s 213 isolates. m . o r 214 Our findings corroborate previous reports demonstrating increased proportions of intrinsically less azole- g / o 215 susceptible species (primarily C. glabrata) among invasive (6) as well as colonizing Candida isolates (7) n J a 216 following azole prophylaxis. In contrast, only discrete and non-significant differences in species distribution n u a 217 were noted for patients treated with echinocandins. Echinocandin prophylaxis was previously r y 2 218 demonstrated to significantly drive the invasive Candida species towards intrinsically less susceptible 9 , 2 219 species (mainly C. parapsilosis) but that was shown in regions where the background prevalence of C. 0 1 9 220 parapsilosis was higher than in the current settings (1, 6, 8, 32, 33). b y g 221 Interestingly, azole treatment resulted in more culture negative and fewer polyfungal post-treatment oral u e s 222 cultures when compared to those from echinocandin exposed patients despite the same median treatment t 223 duration of 14 days. As pretreatment oral swabs could not be obtained for logistic reasons we cannot rule 224 out that more patients treated with fluconazole theoretically might have been culture negative in the oral 225 cavity before treatment and thus that subsequent differences were not related to the choice of antifungal