Breakthrough Aspergillus fumigatus and Candida albicans double infection during caspofungin treatment: laboratory characteristics and implication for susceptibility testing

Abstract

Caspofungin is used for the treatment of acute invasive candidiasis and as salvage treatment for invasive aspergillosis. We report characteristics of isolates of Candida albicans and Aspergillus fumigatus detected in a patient with breakthrough infection complicating severe gastrointestinal surgery and evaluate the capability of susceptibility methods to identify candin resistance. The susceptibility of C. albicans to caspofungin and anidulafungin was investigated by Etest, microdilution (European Committee on Antibiotic Susceptibility Testing [EUCAST] and CLSI), disk diffusion, agar dilution, and FKS1 sequencing and in a mouse model. Tissue was examined by immunohistochemistry, PCR, and sequencing for the presence of A. fumigatus and resistance mutations. The MICs for the C. albicans isolate were as follows: >32 microg/ml caspofungin and 0.5 microg/ml anidulafungin by Etest, 2 microg/ml caspofungin and 0.125 microg/ml anidulafungin by EUCAST methods, and 1 microg/ml caspofungin and 0.5 microg/ml anidulafungin by CLSI methods. Sequencing of the FKS1 gene revealed a mutation leading to an S645P substitution. Caspofungin and anidulafungin failed to reduce kidney CFU counts in animals inoculated with this isolate (P > 0.05 compared to untreated control animals), while both candins completely sterilized the kidneys in animals infected with a control isolate. Disk diffusion and agar dilution methods clearly separated the two isolates. Immunohistochemistry and sequencing confirmed the presence of A. fumigatus without FSK1 resistance mutations in liver and lung tissues. Breakthrough disseminated aspergillosis and candidiasis developed despite an absence of characteristic FKS1 resistance mutations in the Aspergillus isolates. EUCAST and CLSI methodology did not separate the candin-resistant clinical isolate from the sensitive control isolate as well as did the Etest and agar methods.

FIG. 1.

Etest susceptibility endpoints for the day 4 candin-susceptible C. albicans S isolate, the day 38 candin-resistant C. albicans R isolate, and a susceptible unrelated control C. albicans C isolate, respectively, to caspofungin (CAS) and anidulafungin (Anid). From left to right, data for C. albicans S and caspofungin Etest, C. albicans S and anidulafungin Etest, C. albicans R and caspofungin Etest, C. albicans R and anidulafungin Etest, C. albicans S and caspofungin Etest, and C. albicans C and anidulafungin Etest are shown. The MICs for each isolate and antifungal are indicated below the photograph.

FIG. 2.

In vivo susceptibility of C. albicans R and C to anidulafungin, caspofungin, and fluconazole in a murine invasive candidiasis model. Kidney burden on day 4 is shown for mice challenged with C. albicans R (solid marks) and C. albicans C (open marks) and subsequently treated with caspofungin (circles), anidulafungin (squares), fluconazole (triangles), or glucose (control group) (diamonds).

FIG. 3.

Agar dilution susceptibility testing. Repeated testing of C. albicans S (a), C. albicans R (b), and the control C. albicans C isolate (c) to anidulafungin and caspofungin. Concentrations (μg/ml) of the antifungals are indicated next to the wells.

FIG. 4.

Sequence alignment corresponding to nucleotides 1837 to 2136 of the FKS1 region (amino acids 613 to 712 of Fks1p) of C. albicans R (resistant) and C (control). The mutated site is indicated by a box.

FIG. 5.

Immunohistochemical staining demonstrating the presence of Aspergillus hyphae in liver tissue. (The fungal elements are stained with anti-Aspergillus antibodies as described in Materials and Methods and appear red.)