One Size Fits All? Application of Susceptible-Dose-Dependent Breakpoints to Pediatric Patients and Laboratory Reporting

TEXT

The Clinical and Laboratory Standards Institute (CLSI) publishes two documents that are widely used by clinical laboratories to inform susceptibility testing practices. M100 (Performance Standards for Antimicrobial Susceptibility Testing) (1) provides recommendations for bacteria and is updated annually, whereas M60 (Performance Standards for Antimicrobial Susceptibility Testing of Yeasts) (2) addresses yeast testing and is updated periodically. Both documents provide standards for test methodology and quality control (QC) as well as MIC and disc diffusion (DD) clinical breakpoints. Updates to the standards occur when new data on microbial resistance mechanisms, pharmacokinetics (PK) (further defined in the supplemental material) and pharmacodynamics (PD), or clinical outcomes suggest that the historical standard no longer predicts therapy outcomes. At an institutional level, collaboration between the microbiology laboratory, pharmacists, and medical providers should define which aspects of these updated standards are adopted and what antimicrobial reporting, patient treatment, and antimicrobial stewardship practices are best suited to the patients served.

Defining antimicrobial therapy is relatively straightforward when in vitro MIC or DD results fall within the resistance or susceptibility categories. However, deciding the best course of action can be challenging when the MIC or DD result falls in the intermediate interpretive category. Many providers view an intermediate result akin to resistance and switch to another drug class that may be more potent but also carries the potential risk of greater adverse drug events. In addition, changing to another drug class is not always the best antimicrobial stewardship practice and may further drive antimicrobial resistance from drug exposure (3, 4). In response to these concerns, CLSI replaced the intermediate category with “susceptible-dose dependent” (SDD) for select drug/microbe combinations. An SDD interpretation is meant to indicate that increased drug exposure above what is possible with typically used doses, through either increased doses, more frequent doses, and/or infusion times, can predict susceptibility.

The SDD categories established to date (Table 1) were developed using PK/PD, toxicodynamic, and clinical outcome data from adult patients with normal renal function (5–9). It is known that antimicrobial PK in pediatric patients differ substantially from what is seen in adult patients. Children have an evolving physiology which results in changing gastric absorption, liver metabolism, renal elimination, and distribution of antimicrobials as they age. These parameters are often substantially different in children than in adults. For the antimicrobials with SDD interpretive categories, enhanced renal clearance in childhood predicts lower overall antimicrobial exposure, even at elevated doses, than in adults. Thus, application of the dosing strategies outlined in CLSI M100 and M60 to target SDD isolates in pediatric patients may not be optimal. Unfortunately, there are limited data to support or refute the use of these SDD breakpoints for children, limiting the applicability of these interpretative criteria. In this commentary, we provide an overview of the antimicrobials with SDD interpretive criteria and explore existing data to evaluate whether the adult SDD category can be extrapolated to the pediatric population.

CEFEPIME

Cefepime is a fourth-generation bactericidal cephalosporin that interferes with peptidoglycan synthesis. In Enterobacteriaceae producing AmpC-type beta-lactamases, cefepime remains active as it is a poor inducer of AmpC expression (10). Furthermore, the zwitterionic properties of cefepime allow faster penetration through the bacterial cell wall than what is possible with other cephalosporins. Cefepime is largely recognized as a carbapenem-sparing option for the treatment of AmpC-producing Enterobacteriaceae and Pseudomonas aeruginosa, often making this a good choice for empirical sepsis therapy (11–13). Cefepime’s bactericidal activity against these Gram-negative bacteria is best predicted by a target serum concentration of unbound cefepime that is greater than the MIC (fT>MIC) (see Table S1 in the supplemental material) for at least 50 to 60% of the dosing interval for respiratory infections and 83 to 95% for infections at other anatomical sites (14, 15). For adults, this target can be achieved with a dose of 1 g every 8 h (q8h) or 2 g q12h for an MIC of 4 μg/ml and of 2 g q8h for an MIC of 8 μg/ml (Table 1). However, outside the neonatal age range, it is very unlikely that a pediatric patient will achieve acceptable cefepime exposures at an MIC of 8 μg/ml, even if high-end pediatric dosing (50 mg/kg q8h) is used with typical infusion rates (e.g., 5 to 30 min). For instance, one study found that the probability of target attainment (PTA) (see Table S1 in the supplemental material) was >90% for term and preterm neonates if the isolate MIC was 4 or 8 μg/ml but that the PTA was only 32% with q12 dosing, 69% with q8 dosing, and 89% with q6 dosing for an MIC of 8 μg/ml for children >30 days of age (16). Courter et al. simulated total cefepime doses of 100 to 150 mg/kg/day with a variety of infusion strategies (30 min, 3 h, or continuous infusion) for children aged 2 to 12 years and found PTAs of >90% for all strategies at an MIC of 4 μg/ml, but only the continuous-infusion strategy achieved an acceptable PTA at an MIC of 8 μg/ml (17).

The differences in the PTAs for neonates versus older children for cefepime relates to altered clearance of the drug in these age groups. One study showed that for neonates <4 months of age (average age, 14 days), cefepime clearance was lower than that for older infants and children, leading to a prolonged serum half-life and 95% PTA for isolates with an MIC of 8 μg/ml when dosed at 30 mg/kg q12h (18). It should be noted that the target fT>MIC was >50% in this study; with a target fT>MIC of 95%, this dose has an ∼85% PTA. Cefepime renal clearance increases as children age, but an fT>MIC of ∼90% was shown to be possible in children aged 2 months to 16 years, in a study of intravenous (i.v.) or intramuscular (i.m.) cefepime PK, for isolates with MICs of ≤4 μg/ml (19).

Clinical outcome studies that evaluate treatment efficacy at SDD MICs are limited to adults, but some recent literature suggests worse treatment outcomes when MICs are in the SDD range for extended-spectrum beta-lactamase (ESBL)-producing isolates (20, 21). Lee et al. (20) concluded from a study in adults that treatment of Enterobacter cloacae ESBL-producing isolates using SDD dosing is associated with a higher 30-day mortality rate than with the use of carbapenems. Klebsiella pneumoniae and Escherichia coli ESBL-producing isolates causing bloodstream infections trended toward a higher mortality if cefepime versus a carbapenem was administered, although this was not statistically significant (22). To date, no data suggest a difference in mortality rate in AmpC, non-ESBL producers with cefepime MICs within the SDD breakpoint, and no outcome studies are available on use of a carbapenem versus cefepime dosed at the SDD concentration.

While the PK data support the use of cefepime in all age groups, different PTAs obtained in various studies, along with the paucity of outcome data, suggest that caution be used when applying the SDD breakpoints in pediatrics. In addition, altered dosing recommendations for neonates with naturally reduced renal clearance should be considered in relation to the SDD breakpoints.

CEFTAROLINE

Ceftaroline is a fifth-generation cephalosporin which, unlike other beta-lactams, possesses activity against methicillin-resistant Staphylococcus aureus (MRSA), as it has affinity for penicillin-binding protein 2a. Like for cefepime, the PK/PD parameter that best predicts efficacy is fT>MIC. In vitro S. aureus models demonstrated that an fT>MIC of ∼35% is required for bactericidal (2-log kill) activity, and this target was used by CLSI to establish the current breakpoints (CLSI June 2018 meeting minutes, rationale document in preparation). CLSI defined MICs of 2 to 4 μg/ml as SDD for S. aureus, with a recommended dosing regimen in adult patients with normal renal function of 600 mg i.v. q8h, administered over 2 h. Although not FDA approved, use of this regimen in adult patients with invasive MRSA infections has been documented throughout the literature (8) and is under investigation by the drug sponsor. It should be noted that some suggest that a target fT>MIC of >50% should be used to avoid emergence of resistance (9, 23).

For ceftaroline, SDD breakpoints are not based on FDA-approved dosing regimens, and neither pediatric nor adult FDA-approved drug dosing regimens are likely to achieve the PK/PD target for isolates in the SDD range, especially the more aggressive target of >50% fT>MIC, which may lead to treatment failure and increased S. aureus drug resistance. Very few data exist to evaluate the safety of off-label ceftaroline doses in pediatrics. One small study evaluated ceftaroline at 15 mg/kg (maximum dose, 600 mg) q8h, infused over a 2-h period, for children >6 months with complicated pneumonia (24). In this study, both efficacy and adverse events were similar to those with the comparator agents, ceftriaxone plus vancomycin. PK modeling of pediatric patients with cystic fibrosis (CF) (ages 6 to 17 years) predicts an fT>MIC of 50% for isolates with MICs of 2 μg/ml for ≥90% of patients if the drug is dosed at 15 mg/kg q8h, infused over a 2-h period. Patients with cystic fibrosis have increased drug clearance and volume distribution, so it is predicted that non-CF patients will have an even higher PTA, although this has not been specifically studied (25). Target attainment would be less probable at the higher SDD MIC of 4 μg/ml. Therefore, if trying to achieve an MIC of 4 μg/ml, clinicians should consider utilizing alternative antibiotics, and if ceftaroline is used, therapeutic drug monitoring, which is currently available only from specialty laboratories (e.g., Atlantic Diagnostics Laboratories, Bensalem, PA), may be helpful, although routine turnaround times may diminish clinical utility in patients decompensating from severe infections (26).

DAPTOMYCIN

Daptomycin is a novel cyclic lipopeptide with rapid Gram-positive bactericidal activity via a calcium-dependent mechanism. Daptomycin retains activity against several resistant Gram-positive pathogens, including MRSA and some vancomycin-resistant Enterococcus faecium and S. aureus strains (27–30). The concentration-dependent bactericidal activity of daptomycin is best described by the pharmacodynamic parameter comparing the area under the dose concentration curve of freely available drug in serum over time to the target MIC (fAUC/MIC) (31). For Enterococcus spp., CLSI estimated the target fAUC/MIC to be >27.4 by performing a post hoc evaluation of published human outcomes data and to be >9.8 from murine neutropenic thigh infection models (32, 33). Monte Carlo simulations (see Table S1 in the supplemental material) demonstrated 2 to 5% PTA at an MIC of 4 μg/ml for the human-derived target and 65 to 84% PTA for the murine-derived target with the current U.S. Food and Drug Administration (FDA)-approved dose of 6 mg/kg/day in adults (34, 35). In contrast, the PTAs at 4 μg/ml were 32 to 54% and 98 to 99% at a dose of 12 mg/kg/day, respectively, using these models. The previous daptomycin susceptibility breakpoint listed in CLSI M100 for all Enterococcus spp. was ≤4 μg/ml. However, these updated PK/PD data and reports of daptomycin treatment failures against E. faecium strains with higher MICs of 4 μg/ml prompted CLSI to revise the susceptible MIC breakpoint for non-E. faecium Enterococcus spp. (from ≤4 μg/ml to ≤2 μg/ml) and to incorporate an SDD category for E. faecium (5, 33). All E. faecium isolates with an MIC of ≤4 μg/ml are now designated SDD based on an adult daptomycin dosing regimen of 8 to 12 mg/kg every 24 h for the treatment of serious E. faecium infection, along with a recommendation for infectious diseases consultation (5, 30, 36–38).

Available pediatric daptomycin pharmacokinetic data highlight an inverse relationship between patient age and daptomycin clearance. Enhanced daptomycin clearance in younger patients results in lower overall drug exposure (i.e., fAUC) (Table S1) than in older children and adults receiving a similar daptomycin dose (29, 39–42). Given the similarities between adolescent and adult daptomycin PK, extrapolation from adult clinical data suggests that adequate PD target attainment may be feasible for patients >12 years of age, although safety data for doses greater than 10 mg/kg are limited (43, 44). Daptomycin monotherapy for the treatment of serious E. faecium infection in children less than 12 years of age is challenging. The available PK literature evaluating total drug AUC exposures for various age groups (<120 days, 3 to 12 months, 13 to 24 months, and 7 to 11 years) at various dosing regimens (6 mg/kg, 4 mg/kg, 10 mg/kg, and 7 mg/kg, respectively) do not correlate with an ƒAUC/MIC of >27.4 at an MIC of 4 μg/ml (40, 43–46). Until better dosing data are available, daptomycin monotherapy for the treatment of serious E. faecium infection in children <12 years of age should be used with caution. Therapeutic drug monitoring and drug exposure calculations should be considered when feasible, particularly in critically ill patients with altered drug clearance and volume of distribution (45).

FLUCONAZOLE

Fluconazole is intended for the treatment of certain fungal infections, including those caused by Candida species, and its mechanism of action is to inhibit cytochrome P450 and prevent ergosterol production used in the cytoplasmic membrane (47). The CLSI M60 SDD range is defined by an MIC of 4 μg/ml for C. albicans, C. parapsilosis, and C. tropicalis and of ≤32 μg/ml for C. glabrata, with the caveat that treatment of C. glabrata infections with fluconazole is appropriate only in select clinical contexts (2). For Candida infections, the PK/PD parameter that best predicts a 50% effective fungistatic dose response is reported to be an fAUC/MIC of 25 to 100 (48, 49).

Adult dosing for SDD infections is recommended at >800 mg/dose, higher than the standard dose of 6 to 12 mg/kg/day used clinically, to achieve at least an total drug exposure in the first 24 h, represented by the AUC_0–24, of 400 mg · h/liter, equivalent to an AUC/MIC of 50, for treatment of Candida spp. with an MIC of ≤8 μg/ml (50, 51). For invasive candidiasis in immunocompromised adults, an AUC_0–24 of 800 mg · h/liter is recommended. A small clinical study of critically ill infants ranging from 6 to 59 days old found that 5 of the 8 achieved an AUC_0–24 of ≥400 mg · h/liter following a one-time 25-mg/kg fluconazole loading dose. Fluconazole clearance was highly variable in this small cohort of patients. None of the infants achieved an AUC_0–24 of 800 mg · h/liter. A Monte Carlo simulation (see Table S1 in the supplemental material) using serum fluconazole concentrations measured from 55 infants age 1 to 90 days old (gestational age of 23 to 40 weeks) suggested that a daily fluconazole dose of 12 mg/kg will achieve a minimum AUC_0–24 of 400 mg · h/liter in at least 90% of infants in this age range, with a median AUC_0–24 exposure of 600 to 800 mg · h/liter (52). In older children, doses of 8 mg/kg/day have been shown to correlate with a mean AUC_0–24 of only 200 mg · h/liter (53), although higher fluconazole doses of 12 mg/kg/day are commonly used in clinical practice. However, even with this higher dose, children (>90 days old) may not achieve the same fluconazole exposure as recommended in adult patients given their rapid renal clearance (54). Given that the CLSI SDD standard is based on the adult doses higher than the standard of 6 mg/kg/day, this suggests that pediatric patients would be unlikely to obtain the required fluconazole concentration necessary for SDD dosing, necessitating utilization of alternative antifungal agents for severe infections.

SUGGESTED LABORATORY COMMENTS

The clinical microbiology laboratory is responsible for testing and reporting antimicrobial susceptibilities for various microorganisms for clinical application. Certain organism-antimicrobial combinations require additional comments as a recommendation for their use or antimicrobial dosage. These comments are especially important for facilities that may not have immediate access to specialty care providers, such as pediatric infectious disease physicians and pharmacists. As such, microbiology laboratories should be prudent to ensure that reported MIC, interpretation by breakpoints, and comments are accurate and clear. Collaboration with the infectious disease department and pharmacy is highly encouraged whenever reporting changes based on updated CLSI guidelines are considered.

For SDD reporting, appendix F of CLSI M100 (29th edition) provides a suggestion for those laboratories wanting to report a comment with the SDD breakpoint. This comment, however, suggests a “one-size-fits-all” dosage regimen without denoting how differences in PK parameters may influence outcomes. We suggest the following comment for consideration for SDD results in addition to the CLSI comment:

“Interpretive criteria for SDD are based on a dosing regimen that optimizes PK/PD targets in adult patients and may not be applicable to pediatric populations. Consultation with a pediatric pharmacist or physician who is familiar with PD targets and pediatric PK is highly recommended.”

APPLICATION TO PRACTICE AND RECOMMENDATIONS

SDD is an important concept for antimicrobial stewardship and treatment management in the era of ever-increasing antimicrobial resistance. However, the data that support use of this interpretive category are largely limited to those derived from adult populations. It is important for clinicians to consider that SDD results are based on adult dosing, the pediatric equivalents of which may or may not be sufficient for all pediatric patients. When concerns regarding sufficient dosing exist, one strategy is to obtain patient-level antimicrobial serum concentrations and to calculate patient-specific pharmacodynamic exposures. This strategy is commonly used for the aminoglycosides and vancomycin, but therapeutic drug monitoring is not routinely performed by many institutions for cefepime, daptomycin, fluconazole, or ceftaroline. Laboratories that serve a large pediatric population should discuss the need for drug monitoring with pediatric clinicians and, if desired, determine whether such testing is possible at reference laboratories, although we were able to identify only one reference laboratory that offers therapeutic drug monitoring for ceftaroline and fluconazole, and none that does so for cefepime or daptomycin, in the United States. If ceftaroline is to be used, up-front work by the laboratory to establish an account and specimen shipment routes will avoid delays in patient management for infections caused by organisms with MICs in the SDD range for this drug.

The data used to inform breakpoint revisions by standards development organizations (SDOs) such as CLSI are often sparse, and pediatric data are often unavailable. We urge that, whenever possible, pediatric-specific doses be considered by SDOs when applying SDD to breakpoints. If these data are not available, there should be a clear indication that dosages used to establish the interpretive criterion are for adult populations alone, as was done by CLSI for daptomycin. If data exist to demonstrate that target attainment for an SDD MIC is not achievable in the pediatric population based on their PK, a statement noting this should be provided. In these instances, laboratories might consider editing SDD results to resistant for pediatric patients to avoid confusion. In all cases, management of pediatric infections caused by isolates with MICs in the SDD range is complicated, requiring evaluation of all data by infectious disease pharmacotherapy experts, such as a pediatric infectious disease pharmacist. Microbiology laboratories may consider adding comments to patient reports that suggest consultation with such specialists. Such comments help translate the CLSI goal of integration of the SDD category and ensure patient safety while optimizing antimicrobial stewardship and positive treatment outcomes.