A Practical Guide to Antifungal Susceptibility Testing

Abstract

We review antifungal susceptibility testing and the development of clinical breakpoints, and detail an approach to using antifungal susceptibility results when breakpoints have not been defined. This information may prove helpful when selecting therapy for invasive fungal infections in children.

INTRODUCTION

Invasive fungal disease (IFD) is a major source of morbidity and mortality, particularly for immunocompromised children [1, 2]. Continued need for chemotherapy regimens to treat cancer, expanding indications for hematopoietic cell transplantation, and increasing utilization of novel immune modulating agents have increased the number of children vulnerable to an IFD [3, 4]. Timely initiation of appropriate treatment is crucial for these infections. While first-line therapies have been determined for the most common fungal pathogens [5–9], the emergence of resistant organisms such as Candida auris, triazole-resistant Aspergillus fumigatus, or fluconazole-resistant Candida species highlights the need for antifungal susceptibility testing [10–15]. Antifungal susceptibility testing (AFST) can provide in vitro measurements of the concentration of an antifungal agent that inhibits growth of a given fungal organism, the minimum inhibitory concentration (MIC) or minimum effective concentration (MEC). For the common pathogens and systemic antifungal agents, clinical breakpoints (CBP) are also available to inform whether the identified MIC/MEC equates to a susceptible or resistant isolate [16–19]. Both the Clinical and Laboratory Standards Institute (CLSI) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) have developed standardized AFST methodologies [20–27]. Commercial assays for AFST are available, and new assays also are being developed. We review indications for performing AFST, discuss common methods of AFST, and provide guidance for interpreting AFST results with and without an accompanying CBP [28, 29].

WHEN SHOULD ANTIFUNGAL SUSCEPTIBILITY BE PERFORMED?

Generally, AFST is indicated for cases of proven or probable IFD, when there is suspicion for acquired resistance, or in cases of refractory or breakthrough disease [16, 28, 29]. AFST is most valuable for organism-drug pairings that have established CBPs. While the utility of AFST for organisms lacking a defined CBP is not well defined, knowledge of in vitro susceptibility to specific antifungals may help guide the clinician in therapeutic decision-making [17].

There are several clinical scenarios when AFST is not likely to be clinically informative. Some fungal organisms have intrinsic resistance to an antifungal agent or class based on functional or structural characteristics [18, 30]. This includes echinocandin resistance in Cryptococcus and Rhodotorula spp. and resistance to fluconazole by Aspergillus spp [18]. Other organisms do not possess intrinsic resistance but may develop resistance to an antifungal agent after a period of exposure. For example, amphotericin B-susceptible Clavispora lusitaniae (formerly Candida lusitaniae) isolates are capable of quickly developing resistance while on therapy [31–33]. In these circumstances AFST is not necessary and the respective antifungal agent should not used. AFST also should not be routinely performed for fungal organisms in cultures obtained from non-sterile sites unless the patient is a risk for fungal disease, as these cultures should not routinely prompt therapy.

ANTIFUNGAL SUSCEPTIBILITY TESTING METHODS

There are several important considerations when using AFST in clinical practice. AFST may be time-consuming, especially for non-Candida spp. Therefore, if a culture from a sterile site results with fungal growth in a patient that is vulnerable to IFD, clinicians should not wait to start therapy but rather initiate empirical therapy per an institution’s local epidemiology. For AFST to be performed, the initial culture must have adequate growth for identification to the genus and species level. Once AFST is requested, it should be recognized that AFST requires labor-intensive phenotypic methods. Although many clinical laboratories will perform AFST for yeasts locally, ASFT for molds is not routinely done at local microbiology laboratories.

Both CLSI and EUCAST have developed standardized AFST procedures for Candida spp. and Aspergillus spp. to limit variability in AFST results between laboratories [20–27]. AFST relies on the in vitro assessment of an organism’s response to an antifungal drug, using controlled conditions which may differ from clinical conditions during treatment of infection. Furthermore, results of in vitro testing can vary based on choice of culture medium, size of inoculum, and incubation characteristics such as temperature or time [16]. Even under these standardized conditions, AFST may not predict therapeutic response for an individual patient, due to clinical factors beyond the specific drug-organism pairing such as a patient’s immune status [20–27].

Broth Microdilution

Broth microdilution involves the use of known concentrations of antifungal agents in liquid culture media [16]. Microdilution plates using standardized media are prepared with two-fold dilutions of an antifungal drug. A defined inoculum of the fungal organism is introduced, and the plate is incubated for at least 24 hours prior to reading. The minimum inhibitory concentration (MIC) for broth microdilution is defined as the concentration of drug that decreases growth of the fungal organism to a prespecified degree. Importantly, the reproducibility of broth microdilution is considered to be plus or minus two doubling dilutions; this contrasts with bacterial susceptibility testing, where the reproducibility is plus or minus one dilution [34].

Both CLSI and EUCAST have developed protocols for broth microdilution for yeasts. There are important differences between the two standards that may impact MIC determination [16, 19]. The organizations use the same culture medium, but different glucose content and inoculum sizes. The physical shape of the plate wells is also different, and the preparation of the antifungal drugs and organism inoculum differ slightly. Notably, in the CLSI protocol wells are read visually with a viewing mirror, whereas in the EUCAST protocol all plates are read using a spectrophotometer microplate reader. The use of the microplate reader reduces subjectivity in endpoint determination [26].

The two yeast AFST protocols have similar criteria to define the endpoint of the AFST assay. For azoles, echinocandins, and flucytosine, the MIC is set at the lowest drug concentration at which there is a ≥50% decrease in growth. For amphotericin B, complete growth inhibition is required by CLSI, defined as 100% decrease for visible growth. EUCAST uses ≥90% decrease in growth as the endpoint as this represents a clear well and prevents background variation from interfering with endpoint definition. However, in some instances organisms will exhibit “trailing growth,” wherein there is growth of residual organism beyond the MIC [16]. This may complicate the reading the results visually and may cause isolates to be misclassified as resistant. However, isolates that exhibit trailing growth are susceptible to the drug.

Broth microdilution protocols also exist for molds, though the process is much more complex [21, 27]. First, the culture needs to support sporulation. After sporulation is achieved, the antifungal agent is added to the culture medium and incubated for a duration of time dependent on the organism. Antifungal agents impact fungal mold growth differentially which requires different endpoints for interpretation. Azoles and amphotericin B inhibit germination. Therefore, the endpoint is lack of growth, with the MIC defined as the first well with complete growth inhibition [22, 23, 35–37]. Echinocandins allow molds to germinate but inhibit growth at the tip of the emerging hyphae. This appears in culture microwell plates as compact, round “rosette” forms instead of extended hyphal growth. The lowest drug concentration that alters growth in this manner is referred to as the minimum effective concentration (MEC).

Differences in methodologies exist between the CLSI and EUCAST standards that may impact determination of the MIC or MEC [16, 19]. These are similar to those previously described for yeasts, including differences in glucose concentration, the shape of wells, inoculum concentration and the fact that EUCAST has validated spectrometer readings for A. fumigatus.

Disk Diffusion

Disk diffusion is a method of AFST that utilizes prepared paper disks with fixed concentrations of antifungal agents [16, 28]. Disks are applied to an inoculated agar and the zone of inhibition is measured after incubation to assess impact on fungal growth. No MIC is provided with disk diffusion, and interpretation of an isolate’s susceptibility is reliant on identifying a CBP from a range of MIC values for that organism-drug pairing that are contained in reference tables. CLSI has an established methodology for disk diffusion testing of Candida spp. [24], which provides zone diameters and interpretative criteria for common Candida spp. and the drugs caspofungin, micafungin, fluconazole, and voriconazole [22, 24]. A CLSI protocol for disk diffusion also has been developed for molds [25, 38], though at this time, there are no interpretive criteria for disk diffusion AFST of molds. Disk diffusion is not utilized by EUCAST due to variable performance in clinical labs during routine testing.

Agar-Based Azole and Echinocandin Resistance Screening

Standardized screening methods have been developed by EUCAST to detect azole-resistant A. fumigatus and echinocandin-resistant Aspergillus species [39–41]. A four-well agar plate is utilized, with one control well and the remaining three wells containing itraconazole (concentration 4.0 mg/L), voriconazole (2.0 mg/L), and posaconazole (0.5 mg/L). Fungal growth in any well besides the control well is considered a positive screen for resistance. Though agar-based azole resistance screening is frequently performed at local microbiology laboratories in Europe, it is not routinely performed by local microbiology laboratories in the United States. The VIPcheck™ azole resistance assay (Mediaproducts BV, Groningen, Netherlands) is commercially available in Europe; there is no similar product available in other regions. For echinocandins, the endpoint for resistance is presence of radiating hyphae from the control and echinocandin-containing wells. Susceptible isolates will yield a disk-like colony morphology without radiating hyphae in echinocandin-containing wells. The agar screening tests are easy to perform and interpret, but positive tests need to be confirmed with broth microdilution per EUCAST protocol [40, 41].

Molecular Testing Methods

In recent years, molecular diagnostic testing modalities have emerged. Unlike existing MIC-based AFST methods that detect phenotypic resistance when an isolate is incubated with an antifungal drug, these novel approaches require knowledge of molecular resistance mechanisms for an organism-drug pairing [42]. For example, mutations in the fks gene in Candida spp. can confer resistance to echinocandins. Various methods have been developed to detect molecular resistance mechanisms and are reviewed in detail elsewhere [42]. Molecular resistance testing methods can only detect known resistance mechanisms and cannot fully replace phenotypic culture-based methods. However, molecular testing can be helpful in regions with limited infrastructure for phenotypic AFST methods or high rates of specific resistance mechanism, when phenotypic methods result with a borderline MIC/MEC value, or if a patient is failing therapy despite receiving appropriate antifungal regimen [42]. As our understanding of resistance mechanisms expands, the utility of molecular resistance testing will also increase.

Commercial Testing Methods

In addition to the CLSI and EUCAST reference methods, several AFST methods are commercially available. It is important to note that laboratory-to-laboratory variation has been identified when using these methods. Any laboratory that chooses to adopt a commercial AFST method needs to perform an in-house validation to ensure that the results mirror those of a reference method. A detailed discussion of these approaches is available elsewhere [16, 19, 28] but three of these methods will be briefly described here.

Gradient Diffusion

Gradient diffusion uses thin strips containing a defined gradient of antifungal drug with a concentration scale. This test is frequently called the E-test based on the commercial product (bioMérieux, Hazelwood, MO, USA), though products from other vendors exist (Liofilchem, Waltham, MA, USA). These strips have been developed for nearly all approved antifungal drugs [43]. After incubation with a culture of a fungal isolate, the point at which the ellipsoid zone of inhibition intersects the concentration scale is defined as the MIC. The performance of gradient diffusion varies significantly across studies, with significant discrepancies reported for certain organism-drug pairings [43].

Sensititre YeastOne

The Sensititre YeastOne (ThermoFisher Scientific, Waltham, MA, USA) is a broth microdilution product composed of a 96-well plate containing serial dilutions of an antifungal drug in a defined medium with a colorimetric indicator [16]. The MIC is determined based on the first well to show a color change compared to the control well. This platform is easy to perform and interpret [19].

Vitek 2 Yeast Panels

This commercial panel (bioMérieux, Hazelwood, MO, USA) offers automated AFST for certain Candida spp [16, 19]. The panel is essentially a miniaturized and automated version of the doubling dilution technique used in broth microdilution. Prepared cards contain serial concentrations of antifungal drugs, including fluconazole, voriconazole, amphotericin B, and echinocandins. After inoculation onto the card, the card is then incubated to allow for fungal growth. Spectrophotometric readings are used to generate MIC values, as well as interpretive categories for each antifungal drug. The process is rapid, allowing for high testing volume, and automated readings remove the subjectivity of interpretation. However, cost may be prohibitive for laboratories that do not perform a high volume of yeast AFST. It also should be noted that this method has been reported to misclassify Candida strains with fks mutations as susceptible to echinocandins [44].

INTERPRETING ANTIFUNGAL SUSCEPTIBILITY TESTING RESULTS

AFST provides an in vitro measure of how a fungal isolate grows after exposure to increasing concentrations of antifungal agents, with the MIC or MEC serving as a measure of antifungal activity [16–18]. MICs and MECs are limited as they only describe an in vitro phenomenon. Linking MIC/MEC data to clinical outcomes, as is done with CBPs, allows for interpretation as to whether treatment with an antifungal agent will be successful for the patient with the IFD. When CBPs are not available, epidemiological cutoff values can inform on of the potential for resistance for a specific isolate. Both approaches are discussed in more detail below.

Antifungal Clinical Breakpoints

A CBP is a threshold value that allows for organisms to be classified as susceptible or resistant to a specific antifungal drug [16, 18]. CBPs are defined by committees such as CLSI, EUCAST, or the United States Food and Drug Administration [22, 23, 35–37]. The process of defining a CBP for an antifungal drug is similar to that for antibiotics [45]. Both CLSI and EUCAST consider multiple factors during CBP development, including the MIC distribution for an organism/drug combination, pharmacokinetic properties of each drug, pharmacodynamic factors, common dosing schema, the defined wild-type population, and prior research correlating patient outcomes with specific MIC values [16, 18].

If the isolate’s MIC/MEC is below the defined susceptibility breakpoint the isolate is considered susceptible (S); if above the defined resistance breakpoint, resistant (R). CLSI provides a third category, “intermediate (I),” which is used as a buffer zone between S and R for MICs that cannot be easily translated to S or R, and a fourth category, “susceptible-dose-dependent (SDD),” which for antifungals is only used for fluconazole against Candida. This indicates that higher doses of an antifungal drug are needed for a drug to be effective. EUCAST has a similar designation, “susceptible, increased exposure” (I), where the organism can be regarded as susceptible provided higher exposure is achieved at the site of infection (through increased dosing or physiologic concentration at the site). EUCAST no longer has an intermediate category. Instead, EUCAST uses the term “area of technical uncertainty” (ATU) when MICs for resistant mutants overlap with the MIC range for susceptible wild-type organisms and the MIC value cannot by itself be classified as susceptible or resistant. The ATU designation is not to be reported to the clinician but comes with guidance on next steps for laboratories (perform additional testing, etc.) as well as instructions for reporting of MICs to clinicians [46].

Antifungal CBPs exist for the most frequently encountered fungal organisms and the antifungal drugs used to treat them [22, 23, 35–37]. Select CBPs are shown in Table 1; these CBPs were current at the time of publication may be reviewed and revised by CLSI and EUCAST as new data become available [48]. There remains a paucity of data linking in vitro AFST results to clinical outcomes for many organisms, which limits the ability of CLSI and EUCAST to define CBPs for rare pathogens or certain organism-drug pairings [16, 19].

Table 1.

Clinical minimum inhibitory concentration breakpoints for select yeasts and molds using EUCAST and CLSI methodologies^a,b

		EUCAST (mg/L)				CLSI (µg/mL)
Species	Drug	S (≤)	Ic	R (>)	ECOFF	S (≤)	Ic	SDD	R (≥)	ECV
Candida albicans	Amphotericin B	1	——	1	1	—	—	—	—	2
	Anidulafungin	0.03	—	0.03	0.03	0.25	0.5	—	1	0.12
	Micafungin	0.016	—	0.016	0.016	0.25	0.5	—	1	0.03
	Rezafungin	—	—	—	—	0.25	—	—	—	0.06
	Fluconazole	2	4	4	0.5	2	—	4	8	0.5
	Itraconazole	0.06	—	0.06	0.03	—	—	—	—	—
	Posaconazole	0.06	—	0.06	0.06	—	—	—	—	0.06
	Voriconazole	0.06	0.125–0.25	0.25	0.03	0.12	0.25–0.5	—	1	0.03
Nakaseomyces glabrata (formerly Candida glabrata)	Amphotericin B	1	—	1	1	—	—	—	—	2
	Anidulafungin	0.06	—	0.06	0.06	0.12	0.25	—	0.5	0.25
	Micafungin	0.03	—	0.03	0.03	0.06	0.12	—	0.25	0.03
	Rezafungin	—	—	—	—	0.5	—	—	—	0.12
	Fluconazole	0.001	≤16	16	16	—	—	≤32	64	8
	Itraconazole	—	—	—	2	—	—	—	—	4
	Posaconazole	—	—	—	1	—	—	—	—	1
	Voriconazole	—	—	—	1	—	—	—	—	0.25
Pichia kudriavzevii (formerly Candida krusei)	Amphotericin B	1	—	1	1	—	—	—	—	2
	Anidulafungin	0.06	—	0.06	0.06	0.25	0.5	—	1	0.25
	Micafungin	—	—	—	0.25	0.25	0.5	—	1	0.25
	Rezafungin	—	—	—	—	0.25	—	—	—	0.12
	Fluconazole	—	—	—	128	—	—	—	—	—
	Itraconazole	—	—	—	1	—	—	—	—	1
	Posaconazole	—	—	—	0.5	—	—	—	—	0.5
	Voriconazole	—	—	—	1	0.5	1	—	2	0.5
Candida parapsilosis	Amphotericin B	1	—	1	1	—	—	—	—	1
	Anidulafungin	4	—	4	4	2	4	—	8	4
	Micafungin	2	—	2	2	2	4	—	8	2
	Rezafungin	—	—	—	—	2	—	—	—	4
	Fluconazole	2	4	4	2	2	—	4	8	2
	Itraconazole	0.125	—	0.125	0.125	—	—	—	—	0.5
	Posaconazole	0.06	—	0.06	0.06	—	—	—	—	0.25
	Voriconazole	0.125	0.25	0.25	0.06	0.12	0.25–0.5	—	1	—
Candida tropicalis	Amphotericin B	1	—	1	1	—	—	—	—	2
	Anidulafungin	0.06	—	0.06	0.06	0.25	0.5	—	1	0.12
	Micafungin	—	—	—	0.06	0.25	0.5	—	1	0.06
	Rezafungin	—	—	—	—	0.25	—	—	—	0.12
	Fluconazole	2	4	4	1	2	—	4	8	1
	Itraconazole	0.125	—	0.125	0.125	—	—	—	—	0.5
	Posaconazole	0.06	—	0.06	0.06	—	—	—	—	0.12
	Voriconazole	0.125	0.25	0.25	0.125	0.12	0.25-0.5	—	1	0.12
Candida dubliniensis	Amphotericin B	1	—	1	0.25	—	—	—	—	0.5
	Anidulafungin	—	—	—	—	—	—	—	—	0.12
	Micafungin	—	—	—	—	—	—	—	—	0.12
	Rezafungin	—	—	—	—	0.12	—	—	—	0.12
	Fluconazole	2	4	4	(0.5)*	—	—	—	—	0.5
	Itraconazole	0.06	—	0.06	0.06	—	—	—	—	0.25
	Posaconazole	0.06	—	0.06	0.06	—	—	—	—	0.12
	Voriconazole	0.06	0.125–0.25	0.25	0.03	—	—	—	—	—
Candida auris	Anidulafungin	—	—	—	—	—	—	—	—	1
	Micafungin	—	—	—	—		—	—	—	0.5
	Rezafungin	—	—	—	—	0.5	—	—	—	0.5
Aspergillus fumigatus	Amphotericin B	1	—	1	1	—	—	—	—	2
	Itraconazole	1	—	1	1	—	—	—	—	1
	Posaconazole	0.125	—	0.25	0.25	—	—	—	—	—
	Voriconazole	1	—	1	1	0.5	1	—	2	1

Abbreviations: EUCAST, European Committee on Antimicrobial Susceptibility Testing; CLSI, Clinical and Laboratory Standards Institute; S, susceptible; SDD, susceptible, dose-dependent; R, resistant; ECOFF, epidemiological cutoff value; ECV, epidemiological cutoff value.

^aCurrent as of December 15, 2022.

^bSee references [22, 23, 36, 47] for additional drug-organism pairings.

^cThe designation “I” for EUCAST indicates “susceptible, increased exposure” for organisms where there is a high likelihood of therapeutic success when exposure to the antifungal agent is increased by adjusting its dosing regimen or when an antifungal achieves a high concentration at the site of infection. The CLSI designation “I” means “intermediate,” which describes isolates that cannot clearly be categorized as susceptible or resistant. The “intermediate” designation indicates that isolates might respond to higher doses of antifungal drugs or if drug concentration in the infected tissue is maximized.

^*Parentheses “()” denote a tentative ECOFF.

Epidemiological Cutoff Values

An epidemiological cutoff value (abbreviated ECV using CLSI terminology and ECOFF using EUCAST terminology) denotes the upper limit MIC/MEC values for wild-type strains of a fungal species [36, 47]. These values are determined based on a visual inspection of and a statistical calculation from the MIC/MEC distribution [48, 49]. Unlike CBPs, which are based not only on MIC distributions but also pharmacokinetic/pharmacodynamic and clinical data, ECV/ECOFFs do not predict clinical response to therapy [47]. ECV/ECOFFs have only been defined for a certain number of organism-drug pairings due to the rarity of certain fungal infections, resulting in limited clinical and pharmacological data for each pairing (Table 1). However, ECV/ECOFFs still may be used to guide clinical decision-making when CBPs are not available [16]. For isolates with MIC/MEC below the ECV/ECOFF, the isolate is likely wild-type, and clinicians can use the published ECV/ECOFFs to assess if achievable antifungal concentrations with standard dosing regimens will exceed the wild-type MICs for an organism isolated in culture. Isolates with an MIC or MEC above the defined ECV/ECOFF may have acquired resistance and may be less likely to respond to therapy [16–18]. Additionally, clinicians can use the published ECV/ECOFFs to assess if achievable antifungal concentrations with standard dosing regimens will exceed the resulted MIC for an organism isolated in culture [16]. However, it is important to recognize that ECV/ECOFFs are not CBPs and do not predict if a patient will have success or failure upon treatment with an antifungal drug.

USING ANTIFUNGAL SUSCEPTIBILITY DATA WHEN CLINICAL BREAKPOINTS OR EPIDEMIOLOGICAL CUTOFF VALUES ARE NOT AVAILABLE

Interpreting MIC or MEC results from AFST for an organism-drug pairing in the absence of a CBP or ECV/ECOFF can be challenging. A pragmatic approach to these scenarios has been published for yeasts without CBPs [17]. This conceptual model can be used to guide clinicians when selecting an antifungal drug for a fungal organism without a defined CBP or ECV/ECOFF.

First, it is essential to determine that the organism is correctly identified to the genus and species level. Misclassification of an organism may lead the laboratory to erroneously conclude that a CBP does or does not exist for a given clinical isolate. For example, Aspergillus lentulus may be mis-identified as A. fumigatus based on phenotypic methods [50, 51]. Though a genetically distinct species from A. fumigatus, A. lentulus is morphologically similar in culture [52]. Additional detection methods, such as thermotolerance tests, may be necessary to accurately identify the organism to the species level especially if there is concern that the identification may be incorrect [53]. Molecular identification methods also can be utilized if phenotypic identification is inconclusive.

When selecting drugs for AFST, the clinician should choose drugs that are expected to have activity against the fungal isolate. Clinicians can then compare the reported MIC or MEC value for the organism-drug pairing to CBPs or ECV/ECOFFs of a related species for which clinical experience exists. Consider the previous example of A. lentulus—AFST results for that organism can reasonably be compared to CBPs for A. fumigatus. If the reported MIC for the organism is higher than known CBPs, than the isolate is likely resistant to that antifungal. This approach relies on the assumption that related species likely have similar pathogenicity, resistance mechanisms, and potential for invasive disease in an appropriate host and that rare species are rather less virulent than their more common counterparts [17]. However, this approach should be used judiciously as it is an extrapolation from another species and not a published CBP.

ADDITIONAL CONSIDERATIONS WHEN MAKING ANTIFUNGAL THERAPY DECISIONS

The clinician must consider other relevant factors in making antifungal therapy decisions, such as the dosing regimen or pharmacokinetic/pharmacokinetic (PK/PD) properties of that agent. Each antifungal drug differs based on dose, formulation, molecule size, chemical structure, lipophilicity, and their metabolism within the human body [54]. These properties impact whether a drug will achieve the necessary concentrations at the site of infection. Even if AFST suggests an organism is susceptible to an antifungal agent, that agent may not achieve adequate concentrations at the site of infection, leading to treatment failure [54, 55]. Surgical resection of difficult-to-treat foci of infection may be needed. Therapeutic drug monitoring should be performed in children when indicated to ensure that the necessary drug levels are achieved [56]. Special consideration should be given to central nervous system infections, as drug penetration varies greatly [54]. For example, echinocandins may not be ideal for treatment of central nervous system infection as they do not appear to achieve adequate concentrations in the cerebrospinal fluid.

When reviewing the PK/PD for a drug, attention should be given to availability of pediatric-specific PK/PD data. PK properties are influenced by age, body composition, and maturation of both renal function and enzymes involved in hepatic clearance [56–58]. Extrapolation of adult doses to pediatric patients can result in under- or over-dosing for children with IFD [56, 58, 59]. For example, voriconazole displays much lower bioavailability in children than in adults, resulting in a need for higher doses in children. Voriconazole is also cleared more rapidly in children, leading to greater variability in drug levels than in adults. Additionally, the drug formulation use may differ for children than adults, which affects absorption and overall exposure to the drug. Posaconazole comes in four different formulations: extended-release tablet, intravenous injection, suspension, and delayed-release powder. Adults frequently receive the tablet form, though children may take the suspension due to inability to swallow pills. However, the formulations are not bioequivalent, meaning interpatient variability in drug levels due to differential absorption is a major issue [56, 60]. More frequent therapeutic drug monitoring may be needed due to variability in drug levels seen with azole antifungals. As such consultation with a clinical pharmacist with expertise in pediatric infectious diseases is recommended for cases of IFD.

Finally, a patient’s medication administration list must be reviewed to assess for possibility of drug-drug interactions [61]. Such interactions can result in under-exposure to antifungal agent that will reduce likelihood of success or over-exposure that will lead to toxicities [56].

CONCLUSIONS

We have provided an overview of AFST including common methods used for this testing, and guidance on how to interpret the results when CBPs are not available. There have been significant advances in AFST methods in the last few decades, helping to optimize therapy for IFD. However, further research in this area is needed. Systematic validation of commercially available susceptibility testing assays is needed so that AFST can be more routinely performed at local clinical laboratories. Effort should be made to develop CBPs for both approved and novel antifungal agents. Finally, future research is needed to improve our knowledge on pediatric specific antifungal PK/PD and our understanding on how differences in PK/PD in children might alter the interpretation of AFST results [29, 62, 63].

Potential conflicts of interest. M.C.A. has, over the past 5 years, received research grants/contract work (paid to the SSI) from Amplyx, Basilea, Cidara, F2G, Gilead, Novabiotics and Scynexis, and speaker honoraria (personal fee) from Astellas, Chiesi, Gilead, MSD, and SEGES. She is the current chairman of the EUCAST-AFST. B.T.F. serves on a data safety monitoring board for Astellas and Allovir. His institution also receives research funding from Merck and Pfizer for research work that his research team performs.