MALDI‐TOF MS can be used for the identification of microorganism species. We have extended its application to a novel assay of Candida albicans susceptibility to fluconazole, based on monitoring modifications of the proteome of yeast cells grown in the presence of varying drug concentrations. The method was accurate, and reliable, and showed full agreement with the Clinical Laboratory Standards Institute's reference method. This proof‐of‐concept demonstration highlights the potential for this approach to test other pathogens.

e Echinocandin drugs are widely used for the treatment of candidemia. Resistance is considered rare, and only a few cases of breakthrough candidiasis in patients receiving echinocandin have been reported worldwide. We report here for the first time a Candida kefyr isolate that acquired echinocandin resistance very rapidly after the initiation of caspofungin treatment for candidemia. We characterized the FKS gene mutation responsible for the resistance via the comparison of isolates sampled before and during treatment.A 61-year-old woman was admitted to the hematological unit at Pitié-Salpêtrière Hospital (Paris, France) for acute myeloid leukemia. The patient was not then receiving and had no history of receiving antifungal drugs. During chemotherapy-induced aplasia, a blood culture sampled at day 1 yielded positive for yeast at day 3. Considering the clinical condition of the patient, echinocandin antifungal therapy was chosen. Treatment by caspofungin was initiated on day 3 (70 mg the first day and then 50 mg/day; patient weight, 60 kg). Analysis of the strain with API 20C (bioMérieux, France) strips and mass spectrometry (Bruker Daltonics, Germany) clearly identified Candida kefyr (teleomorph, Kluyveromyces marxianus). Before the determination of the MICs, caspofungin treatment was left unchanged. The MICs for caspofungin, micafungin, and anidulafungin as determined by Etest (bioMérieux) were 0.06, 0.03, and 0.03 g/ml, respectively. For fluconazole, the MIC was 0.25 g/ml. As the patient's condition did not improve, treatment was switched to fluconazole at day 10. A blood culture sampled the same day was positive for yeast at day 11. Surprisingly, this culture gave C. kefyr again. For this isolate, the MICs for caspofungin, micafungin, and anidulafungin by Etest were 2, 4, and 8 g/ml, respectively. It should be noted that the determination of MIC by the CLSI reference method showed similar results (Table 1), but considering the recently proposed epidemiological cutoff value (ECV) for echinocandins against C. kefyr (1), for the initial isolate with low MIC echinocandin values, the MIC for caspofungin was above the ECV, whereas those for anidulafungin and micafungin were below. While receiving fluconazole, a blood culture sampled at day 19 was positive for yeast at day 21, and this time Candida krusei was identified. Treatment was then switched to voriconazole at day 23. The patient died at day 30. Of note, the two candidemias were surely of digestive origin; a stool sample collected a few days before the first candidemia was positive for both C. kefyr and C. krusei.Methods, results, and discussion. Faced with both microbiological resistance and clinical therapeutic failure, we decided to compare the hot spots (HS) of the FKS gene sequences of the two C. kefyr strains, the first identified in the blood culture sampled at day 1 (low echinocandin MIC) and the second at day 10 during caspofungin treatment (high echinocandin MIC). Resistance to echinocandin consistently involves these short FKS sequences that code f...

Taken together, these observations suggest an original mechanism conferring resistance to azoles mediated by cyp51A of environmental origin. This uncommon susceptibility pattern might represent a 'missing link' between the wild-type A. fumigatus and the fully azole-resistant strain harbouring the TR46/Y121F/T289A mutations.