Prevalence and Susceptibility Profile of Candida metapsilosis and Candida orthopsilosis: Results from Population-Based Surveillance of Candidemia in Spain

Abstract

We describe the prevalences and susceptibility profiles of two recently described species, Candida metapsilosis and Candida orthopsilosis, related to Candida parapsilosis in candidemia. The prevalences of these species (1.7% for C. metapsilosis and 1.4% for C. orthopsilosis) are significant. Differences observed in their susceptibility profiles could have therapeutic importance.

Candida parapsilosis has emerged as an important nosocomial pathogen in Spain, and infection usually occurs in neonates, transplant recipients, and patients who have received prior antifungal therapy and parenteral nutrition (1). Because of its variable genetic composition, two new species, named Candida metapsilosis and Candida orthopsilosis, have been recently proposed (16).

The interest in a better characterization of the -psilosis group relies not only on the epidemiology of C. parapsilosis but also on the suspected differences in the antifungal susceptibility profiles of C. metapsilosis and C. orthopsilosis. Antifungal compounds such as echinocandins show great variability among C. parapsilosis isolates (12, 13). However, the current literature contains few data describing the drug susceptibilities of these new Candida species (9, 18, 19). We decided to describe the occurrences of these new species among Candida isolates collected from blood samples in an active population-based surveillance of candidemia. We also examined the antifungal susceptibility profiles of those species.

(This study was presented in part at the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, 2007.)

Strains originally identified as C. parapsilosis from a prospective population-based surveillance study for candidemia conducted in Barcelona, Spain, between 2002 and 2003 (2) were included in the study. Isolate identification was confirmed in our laboratory by physiological tests and molecular methods.

For molecular identification, genomic DNA was directly prepared from a single yeast colony (10). DNA segments comprising the internal transcribed spacer 1 (ITS1) and ITS2 regions were amplified and sequenced using universal primers (20). Further analysis was performed by comparison with C. orthopsilosis ATCC 96139, C. metapsilosis ATCC 96144, and C. parapsilosis ATCC 22019 ITS sequences and with the sequences included in the database of the Department of Mycology of the Spanish National Center for Microbiology. Analyses were conducted with InfoQuest FP 4.50 software (Bio-Rad Laboratories, Madrid, Spain), using a maximum-parsimony clustering methodology. Phylograms were rooted to outgroup Candida kefyr ATCC 28838.

The in vitro antifungal susceptibilities of the strains were tested according to a reference microdilution method (AFST-EUCAST, definitive document 7.1) (14). The antifungal agents used were amphotericin B (AMB), flucytosine, fluconazole, itraconazole, voriconazole, ravuconazole, posaconazole, caspofungin, micafungin, and anidulafungin. Statistical differences between MICs were assessed using analysis of variance. P values of <0.01 were considered significant.

Seventy-eight episodes of C. parapsilosis fungemia, representing 23% of all candidemia episodes (a total of 345 cases of Candida bloodstream infections), were first identified. Based on molecular criteria, 6 strains were reidentified as C. metapsilosis, 5 as C. orthopsilosis, and 67 as C. parapsilosis, representing 1.7% (6/345), 1.4% (5/345), and 19.42% (67/345), respectively, of all candidemia cases. Eight patients had a recurrent episode (one with three sequential isolates and seven with two isolates), resulting in 87 total strains analyzed. Based on these results, C. metapsilosis accounted for 6.9% (6/87) of all C. parapsilosis fungemia cases and C. orthopsilosis for 5.7% (5/87).

Figure 1 shows the rooted cladogram of the isolates and the ATCC strains based on the ITS sequences. Three different groups were found. Two of them, with high levels of identity between sequences (near 100%), included C. parapsilosis and C. orthopsilosis sequences. The third group, which included C. metapsilosis isolates, yielded a more heterogeneous group.

FIG. 1.

Phylogenetic tree of the subset of isolates included in the study, obtained by using maximum-parsimony phylogenetic analyses based on ITS sequences. In order to facilitate the visualization of the cladogram, some isolates of C. parapsilosis were not included, but the species distribution and the bootstrap values were identical.

Table 1 shows antifungal susceptibility testing results, including geometric means (GMs), MIC ranges, and MIC₅₀ values for all the antifungal agents tested. The new species were highly susceptible to all the antifungals, particularly to echinocandin compounds (MICs two- to threefold lower) and AMB, compared with C. parapsilosis (P < 0.01). These differences were especially notable for anidulafungin.

TABLE 1.

GMs of MICs and MIC ranges for the isolates included in the study

Antifungal agent	Result for isolates of indicated species (no. of isolates)
	C. parapsilosis (76)			C. orthopsilosis (5)			C. metapsilosis (6)
	GM (mg/liter)	MIC range	MIC50	GM (mg/liter)	MIC range	MIC50	GM (mg/liter)	MIC range	MIC50
AMB	0.14	0.06-0.5	0.12	0.06	0.03-0.12	0.06	0.08	0.06-0.12	0.09
Flucytosine	0.15	0.12-1	0.12	0.12	0.12	0.12	0.14	0.12-0.25	0.12
Fluconazole	0.37	0.12-64	0.5	0.5	0.5	0.5	1.26	0.5-8	1
Itraconazole	0.02	0.02-0.25	0.03	0.04	0.03-0.06	0.03	0.04	0.02-0.06	0.04
Voriconazole	0.02	0.02-0.5	0.02	0.03	0.02-0.03	0.03	0.03	0.02-0.12	0.03
Ravuconazole	0.01	0.02-0.03	0.02	0.02	0.02-0.03	0.02	0.02	0.02	0.02
Posaconazole	0.02	0.02-0.06	0.02	0.03	0.03-0.06	0.03	0.02	0.02-0.03	0.02
Caspofungin	0.40	0.12-1	0.5	0.16	0.12-0.25	0.12	0.22	0.12-0.5	0.25
Micafungin	0.73	0.5-2	1	0.25	0.25	0.25	0.28	0.12-1	0.25
Anidulafungin	0.88	0.25-4	1	0.38	0.25-0.5	0.5	0.17	0.06-1	0.18

Two out 87 (2.3%) isolates exhibited decreased susceptibility to fluconazole (MIC > 4 mg/liter). The overall activities of triazole agents were similar for the three species: most of the isolates were inhibited by ≤0.125 mg/liter of itraconazole or voriconazole and by ≤0.06 mg/liter of ravuconazole or posaconazole.

This report provides evidence that C. metapsilosis and C. orthopsilosis may behave as human pathogens. This population-based description of Candida bloodstream infection states exceptional information on the epidemiology of candidemia in Spain (2). In our area, the overall prevalences of these new species appear to be important, representing between the fifth and sixth most common species (1.7% and 1.4% for C. metapsilosis and C. orthopsilosis, respectively).

Interestingly, C. metapsilosis has been reported to be rarely recovered from clinical samples (6, 7, 16, 19) and to be as lacking in virulence in vitro as C. orthopsilosis and C. parapsilosis (5). C. orthopsilosis, however, was recovered from blood, nails, skin, lungs, urine, and indwelling catheters, highlighting the clinical relevance of this species (17).

The genetic variability observed among the 87 isolates studied confirms the extensive diversity of the ITS region in C. parapsilosis (3, 15, 16). Our results agree with these reported data, showing three easily separable groups of sequences (Fig. 1). The higher sequence dissimilarities were found among C. metapsilosis isolates. Recently, Iida et al. (6) described a different phylogenetic group based on their different ITS1 sequences and a third new species was proposed.

Although molecular methods are not routinely available, the identification of these new species is easily performed by ITS sequencing. This method can be recommended as the alternative for species characterization since biochemical or morphological methods are not discriminatory enough.

Finally, our antifungal susceptibility results for Candida parapsilosis strains are consistent with other studies (4). The strains tested showed high susceptibilities to AMB and new triazoles and low levels of fluconazole resistance. Echinocandin MICs are comparable to those previously reported (12, 13) showing caspofungin slightly more active than micafungin and anidulafungin against C. parapsilosis.

In addition, results for C. metapsilosis and C. orthopsilosis suggest a trend toward a high level of susceptibility to all the agents tested. Major differences between these new species and C. parapsilosis were found in AMB and echinocandin MICs, although the low number of isolates tested makes it difficult to determine any significant conclusion. Tavanti et al. (17) reported a collection of 33 C. orthopsilosis isolates highly susceptible to the antifungals most commonly used and also caspofungin. Others, such as Melo et al., describe differences among echinocandin susceptibilities on biofilms, showing C. parapsilosis as more resistant than C. orthopsilosis and C. metapsilosis (11). Similarly, lower AMB MICs have been observed in C. metapsilosis than in C. parapsilosis (7, 8).

In conclusion, the slight species-specific differences in the activities of some antifungals could have clinical relevance. These results emphasize the need for surveillance programs to include accurate species identification. In fact, these results encourage us to pursue further investigation addressed toward defining more extensively the epidemiology and antifungal profiles of these new species.