The Pharmacokinetics and Safety of Micafungin, a Novel Echinocandin, in Premature Infants

Abstract

Background: 

Candidal fungal infection rates in neonates are increasing and are a significant cause of mortality, especially in low birth weight infants. Micafungin is an echinocandin that works by inhibiting 1,3-β-D-glucan synthase, an enzyme responsible for fungal cell wall synthesis. The objective of this study was to determine the safety and pharmacokinetics of micafungin in premature infants.

Methods: 

This was a phase I, single-dose, multicenter, open-label, sequential-dose trial of intravenous micafungin investigating 3 doses (0.75 mg/kg, 1.5 mg/kg and 3.0 mg/kg) in 18 premature infants weighing >1000 g (n = 6 in each dosage group). A further 5 infants (500–1000 g) were enrolled in the 0.75 mg/kg dosage group only.

Results: 

The mean ± standard deviation gestational age in the >1000 g dosage group was 26.4 ± 2.4 weeks and, on entry, patients had one or more of a variety of underlying conditions, including sepsis, pneumonia and other infections caused by Candida or other species. Micafungin pharmacokinetics in preterm infants appears linear. However, premature infants >1000 g on average displayed a shorter half-life (8 hours) and a more rapid rate of clearance (approximately 39 mL/h per kg) compared with published data in older children and adults. All doses of micafungin were well tolerated and no serious drug-related adverse events were observed.

Conclusions: 

Single doses of micafungin, ranging up to 3.0 mg/kg, appear well tolerated in premature infants weighing >1000 g. The drug's elimination half-life and total plasma clearance in preterm infants appear dissimilar to published values for these parameters in older children and adults. The reason(s) for this apparent difference remain to be investigated.

Infections caused by Candida species are associated with a high rate of mortality (25–54%) in neonates, especially low birth weight infants.¹ The incidence of such infections in this population has recently increased dramatically with some institutions showing as much as a 20-fold increase.^2,3

The echinocandins are a new class of antifungals that act by selectively and noncompetitively inhibiting 1,3-β-D-glucan synthase, an enzyme responsible for fungal cell wall synthesis.⁴ Because this enzyme is not found in mammalian cells, this unique mechanism of action has resulted in a favorable toxicity profile in humans versus current antifungals and has prompted the use of echinocandins in combination with other antifungals.^5–7

Micafungin (FK463; Astellas Pharma US, Inc., formerly Fujisawa Healthcare, Inc., Deerfield, IL) is an echinocandin with demonstrated broad-spectrum fungicidal activity against Candida species (including Candida albicans species, azole-resistant C. albicans species, and non-albicans species, including Candida krusei, Candida glabrata, Candida tropicalis and Candida parapsilosis) both in vitro^8,9 and in vivo.¹⁰ The pharmacokinetic, efficacy and safety profile of micafungin has been studied in a variety of adult and pediatric patient populations.^6,11–16 To date, the disposition characteristics of this drug have not been evaluated in neonates. Thus, the objective of this study was to evaluate the pharmacokinetics, safety and tolerance of a single dose (3 dose levels) of micafungin in premature infants.

MATERIALS AND METHODS

This was a phase I, single-dose, multicenter, open-label, sequential-dose trial. The study protocol was reviewed and approved by the Institutional Review Board at the participating centers and written informed consent (parental permission) was obtained from the parents or legal guardians for each patient before initiation of any study-related procedures.

Patient Population.

Study patients were premature infants ≤40 weeks postconceptional age who were receiving systemic antifungal therapy. The planned sample size was a total of 18 patients with body weight >1000 g (6 to be enrolled at each dose level). An additional 5 preterm infants weighing 500 g to 1000 g were enrolled into the 0.75 mg/kg dose level. However, further enrollment of patients ≤1000 g was not successful as a result of slow enrollment and technical limitations associated with the demands of the pharmacokinetic study.

Patients who were not eligible for the study included those with biochemical evidence consistent with possible hepatocellular dysfunction (defined as alanine aminotransferase >3 times the upper limit of normal for age), unstable renal function (reflected by continued fluctuations in the serum creatinine of >0.2 mg/dL change daily), hemodynamic instability (requiring pressor support), clinically significant anemia or a concomitant medical condition that increased risk to the patient and/or complicated performance of study-related procedures.

Assessments.

The following observations and measurements were assessed at baseline before micafungin administration: physical examination (including weight and vital signs), hematology and serum chemistry laboratory evaluations (and again at 48 hours after micafungin administration), evaluation of concomitant medications (24 hours before, during and within 72 hours after micafungin administration), demographics and pertinent medical history.

Drug Administration.

In previous maximum tolerated dose studies, the maximum tolerated doses for micafungin were not reached even at the highest doses tested, indicating that doses up to 8.0 mg/kg per day in adults¹⁶ and up to 4.0 mg/kg per day in pediatric patients¹⁴ could be administered safely. This study was designed to begin at a dose of 0.75 mg/kg with escalation to the 1.5 mg/kg and 3.0 mg/kg dose groups in the absence of any apparent drug-associated adverse events as assessed independently by the investigator and study medical monitor. All patients received the single micafungin dose through a syringe pump and microbore intravenous tubing over a 30-minute interval.

Sample Collection, Processing and Analysis.

Blood specimens for quantitation of micafungin (0.15 mL per sample) were obtained from an indwelling vascular catheter not used for study drug infusion or by heel stick immediately before micafungin infusion (time zero); at the end of infusion (30 minutes); and at 2, 8, 12 and 24 hours after the start of the micafungin infusion. Blood samples were transferred into glass vials containing sodium heparin. After mixing by inversion, samples were maintained at 4°C until processing, a period that did not exceed 72 hours from the time of collection. Plasma was separated by centrifugation (1000 rpm for 10 minutes at 4°C), removed by manual aspiration and immediately frozen at −70°C until analysis by a contract laboratory (MDS Harris Laboratories, Lincoln, NE).

A validated high-performance liquid chromatography method with fluorescence detection was used for the quantitation of micafungin from plasma. In brief, plasma samples were acidified with phosphoric acid, precipitated with acetonitrile and centrifuged before dilution with buffer followed by direct injection. Quantitation was achieved in each sample using peak area ratios for micafungin and the internal standard (0.25 μg/mL FR195743). The linear range for micafungin from plasma was 0.05 μg/mL to 25 μg/mL with a limit of quantitation of 0.05 μg/mL. The intra- and interassay coefficients of variation were 3.69% to 7.14%, respectively, over a range of micafungin concentrations of 0.05 μg/mL to 25 μg/mL. All individual sample concentration data reported for micafungin were within the linear range for the assay.

Pharmacokinetic Analysis.

The pharmacokinetic profile of micafungin was determined from the concentration-time data using noncompartmental methods¹⁷ using WinNonlin versions 3.1 and 4.0.1 (Pharsight Corp., Mountain View, CA). The maximum concentration (C_max) and the time of maximum concentration were obtained directly from the plasma concentration-versus-time data. The area under the concentration-time curve (AUC) was calculated from time of dosing (0 hours) until 24 hours (AUC_0–24) by numeric integration using the linear trapezoidal rule for ascending concentrations and the log trapezoidal rule for descending concentrations. AUC to infinity (AUC_inf) was estimated as AUC_0–t + C_t/k_e, where C_t is the last measurable concentration and k_e is the apparent terminal elimination rate constant obtained by log-linear regression of the terminal phase data. Total plasma clearance (Cl) was calculated as dose/AUC_inf. Similarly, the steady-state apparent volume of distribution (Vd_ss) was calculated as (dose)(area under the first moment curve)/(AUC)². The elimination half-life (t_1/2) was calculated as ln(2)/k_e.

Statistical Analysis.

Outliers were assessed using a method described by Tukey.¹⁸ The relationships, if any, between micafungin disposition and patient demographics, including age (gestational [conception to birth], postnatal [after birth] and postconceptional [onset of pregnancy/after implantation of blastocyst] ages), weight and major organ function were assessed by linear regression and r² calculated in Excel 97 (Microsoft Corp., Redmond, WA). Differences in weight-adjusted Cl between the neonatal group and the 2- to 8-year-old, 9- to 17-year-old and adult groups were assessed using the t test function in Excel 97. The Bonferroni multiple-comparison-adjusted level of statistical significance to assess weight-adjusted Cl differences was α = 0.0167 (0.05/3) to maintain an overall α = 0.05 for the 3 comparisons.

Safety/Tolerance Evaluation.

Tolerance was evaluated by physical examination, measurement of vital signs and clinical laboratory assessments (as outlined previously) before and within 48 hours of the single micafungin intravenous infusion. Baseline conditions and adverse events that occurred during micafungin infusion and within 72 hours postinfusion were reviewed by investigators. Investigators recorded whether they thought a given adverse effect was study drug-related.

RESULTS

Patient Demographics.

For the >1000 g weight group, a total of 18 patients were enrolled (6 in each dosage) and all received the full dose of the study drug. The mean ± standard deviation (SD) duration of infusion was 30.8 ± 2.6 minutes (range, 28–40 minutes). Demographic data for the >1000 g weight group are provided in Table 1. Overall, the majority (72.2% [13 of 18]) of patients were white and 66.7% (12 of 18) were male. The mean ± SD gestational age was 26.4 ± 2.4 weeks and the mean baseline weight was 1497.4 ± 303.3 g. On entry, patients had one or more underlying conditions, including sepsis (13 of 18), pneumonia (3 of 18) and other infections (8 of 18) caused by Candida or other species.

TABLE 1: Patient Demographics and Baseline Characteristics of the >1000 g Weight Group (N = 18)

The 5 patients in the 500 g to 1000 g weight group were 60% (3 of 5) female and 80% (4 of 5) black. The mean gestational age was 25.6 ± 1.3 weeks, the mean baseline weight in the 500 g to 1000 g group was 847.6 ± 173.8 g and, on entry, patients had one or more underlying conditions, including sepsis (3 of 5), pneumonia (2 of 5) and other infections (2 of 5) caused by Candida or other species.

Pharmacokinetics Results for the >1000 g Weight Group.

One patient (in the 1.5 mg/kg dose group) had no appreciable change in micafungin plasma concentrations from 8 to 24 hours and, consequently, K_e, Cl, t_1/2, mean residence time in the body (MRT_inf) and Vd_ss could not be estimated for this patient; therefore, this patient was not included in the descriptive or statistical analysis. The mean ± SD plasma concentrations for all patients for each dose group (excluding the one patient from the 1.5 mg/kg dose group mentioned previously) are illustrated in Figure 1. For each dose group, the mean plasma concentration-versus-time curve appeared to be biphasic.

FIGURE 1.: Mean micafungin plasma concentrations ± standard deviation for 6 premature infants (weighing >1000 g) in each dose group (excluding one patient from the 1.5 mg/kg dose group whose terminal elimination rate was indeterminate).

Pharmacokinetic values were calculated for the 15 patients (all with a body weight >1000 g) who had complete concentration-versus-time data to enable accurate determination of pharmacokinetic parameters (summarized in Table 2). As reflected by the mean values for both C_max and AUC_0–24, relative dose proportionality was evident for micafungin. As well, apparent elimination t_1/2, MRT_inf and apparent Vd_ss appeared to be independent of dose.

TABLE 2: Mean ± SD (range) Micafungin Pharmacokinetic Parameters Relative to Dose Administered in the >1000 g Weight Group

When micafungin pharmacokinetic data were compared between dose groups (Table 2), no statistically significant differences were found. Variability was evident in the apparent total plasma Cl for micafungin and was greatest in the 0.75 mg/kg dose group (ie, Cl = 39.0 ± 18.0 mL/h/kg). When micafungin Cl, t_1/2, MRT_inf and Vd_ss were examined as functions of gestational, postnatal or postconceptional age, no significant linear or nonlinear relationships were identified (data not shown). Similarly, none of the micafungin pharmacokinetic parameters differed between sexes or race (ie, white versus black).

Pharmacokinetic Results for the 500 g to 1000 g Weight Group.

Of the 5 infants weighing 500 g to 1000 g that were enrolled, an insufficient volume of blood was drawn to permit determination of the micafungin plasma concentrations for one of these infants, leaving only 4 infants with evaluable pharmacokinetic results. The pharmacokinetics of this smaller weight group was similar to that of the >1000 g group except that the smaller weight group had a shorter mean t_1/2 (5.5 hours) and a more rapid mean Cl per body weight (79.3 ± 12.5 mL/h/kg). Because the number of infants in this weight group was so low, the pharmacokinetic data from this weight group are limited and have not been included in the discussion of this article.

Micafungin Safety.

All doses of micafungin were well tolerated, and there were no treatment-emergent adverse events leading to study discontinuation and no instances of mortality in this study. Most adverse events were mild to moderate in intensity. Two serious adverse events were reported: one patient (>1000 g) experienced severe necrotizing enterocolitis and another patient (>1000 g) experienced moderately increased respiratory distress as a result of accidental extubation as well as life-threatening bronchopulmonary dysplasia. Neither of these events was considered to be related to micafungin administration. There was only one adverse event considered possibly related to the study drug: moderate hypokalemia occurring in one patient (>1000 g) in the 3.0 mg/kg dose group (serum potassium concentration of 4.0 mmol/L at baseline/day 1 and 2.5 mmol/L on day 3). The hypokalemia was persistent at the end of the study. There were no other clinically significant changes in laboratory values related to micafungin.

DISCUSSION

The prevention and treatment of neonatal fungal infections is complex^19–21 and, currently, there are no antifungal drugs approved by the U.S. Food and Drug Administration for use in neonates. Although published clinical experiences in neonates exist for amphotericin B, ^22–24 amphotericin B liposomal formulations^22,24–26 and fluconazole,^23,24,27,28 these standard treatments are often hindered by drug-associated toxicity or development of resistant strains.^19,20 Similarly, pharmacokinetic and safety data in neonates are sparse and limited to only amphotericin B,^24,29,30 amphotericin B lipid complex,³¹ fluconazole^24,32,33 and 5-flucytosine.²⁴

The pharmacokinetics of micafungin have been previously studied in healthy adult volunteers,^34–37 adult patients undergoing bone marrow or peripheral stem cell transplant,^6,16,38 adult patients with renal³⁷ and hepatic impairment³⁷ and pediatric patients with neutropenia.¹⁴ Micafungin pharmacokinetics in both adult and pediatric patients appears linear and follows a 2-compartmental model with rapid distribution into tissues followed by a slow elimination phase^6,14 with steady-state concentrations occurring by day 4 of once-daily dosing.^14,34 Micafungin volume of distribution is small, approximating 0.3 L/kg in pediatric patients¹⁴ and adults.⁶ Micafungin is highly plasma protein-bound (>99%), primarily to albumin. However, because micafungin binding is noncompetitive and does not displace bilirubin binding to albumin, it would be anticipated that micafungin would not pose a risk of kernicterus in neonates.³⁹ Metabolism occurs mainly in the liver through both noncytochrome P450 and cytochrome P450 pathways. Fecal excretion is the major route of elimination.

As recently reviewed by Kearns et al,⁴⁰ developmental differences exist in the disposition of many drugs as evidenced by comparison of pharmacokinetic values for a given drug among neonates, infants, children, adolescents and adults. We explored the potential for age-associated alterations in micafungin pharmacokinetics by comparing the data from our neonatal cohort (>1000 g, n = 15) with those previously reported from populations of children (2–8 years old, n = 33, and 9–17 years old, n = 32; all with fever and neutropenia)¹⁴ and adults (n = 48).⁶ The aforementioned 3 studies were comparable with respect to single-dose administration, the analytical method used to quantitate micafungin from plasma and the approach used for the estimation of pharmacokinetic parameters.

Pharmacokinetic data for micafungin from our neonatal cohort (>1000 g) and those previously reported for children¹⁴ and adults⁶ are summarized in Table 3. Despite significant intersubject variability reflected by wide and overlapping 95% confidence intervals, the mean Cl for micafungin in neonates >1000 g (38.9 mL/h per kg) was approximately 1.7-fold and 2.6-fold greater compared with the value for the Cl reported previously¹⁴ from 2- to 8-year-old children and 9- to 17-year-old children, respectively (P < 0.0167 for each comparison using a 2-tailed test with unequal variance with Bonferroni adjustment for multiplicity). An apparent inverse relationship between micafungin Cl and age was apparent. Similarly, comparison of the mean values for K_e and Vd_ss between neonates >1000 g and children 2 to 8 years of age revealed significant differences for both parameters.

TABLE 3: Comparison of Micafungin Pharmacokinetic Parameters Between the Present Neonate Cohort and Populations of Children and Adults

In adults,⁶ the mean Vd_ss for micafungin (0.26 L/kg) approximates the extracellular fluid space. The apparent larger Vd_ss found for neonates >1000 g and younger children compared with that for older children and adults was expected given that the extracellular fluid space (expressed per unit of body mass relative to the total body water space) is greatest in neonates, infants and young children compared with that for older children and adults.⁴⁰ Given that the functional activity of many of the enzymes apparently responsible for micafungin biotransformation (eg, CYP1A2, CYP2D6, CYP2C, CYP3A4) is markedly lower in neonates compared with older children and adults,⁴¹ the finding of an apparent increase in micafungin Cl in the youngest, most immature subjects was an unexpected finding. Further study of the possible explanations for this apparent increase in micafungin elimination in neonates is required and should include exploration of alternate pathways of metabolism, the ability of CYP3A5 to catalyze the biotransformation of micafungin and the potential impact of age-associated alterations in the drug's degree of plasma protein binding⁴⁰ on micafungin Cl.

Despite the apparent developmental dependence in micafungin disposition, the plasma concentration-versus-time profile and dose proportionality observed in our single-dose neonatal study can be used to examine possible exposure-response relationships in the context of drug treatment. AUC comparisons suggest that doses of 5 mg/kg to 7 mg/kg in neonates >1000 g will approximate the AUC drug exposure of adults receiving daily doses of 100 mg and 150 mg as used to treat invasive candidiasis⁴² and esophageal candidiasis.¹² To further define the appropriate dosage of micafungin in neonatal candidiasis, a preclinical evaluation of micafungin is ongoing for the treatment of Candida meningoencephalitis.

ACKNOWLEDGMENTS

The authors thank Leona C. Fernandes, MSc, who provided medical writing services on behalf of Astellas Pharma US, Inc.

Funding provided by Astellas Pharma US, Inc. (formerly Fujisawa Healthcare, Inc.), Deerfield, IL, through grants to the participating sites (all) and in part by grant #5U10HD031323-13 (M.D.R. and J.L.B.), grant #5U10HD045993-02 (J.N.A.), and grant #5U10HD01313-11 (G.L.K.), Network of Pediatric Pharmacology Research Units, National Institutes of Health/National Institute for Child Health and Human Development (NIH/NICHD), Bethesda, MD.

Portions of this work were previously presented in abstract form: Heresi GP, Gerstmann DR, Blumer JL, et al. Pharmacokinetic, safety, and tolerance study of micafungin (FK463) in premature infants [Abstract 1808]. In: Abstracts of the Pediatric Academic Societies' annual meeting, Seattle, WA, May 3–6, 2003. Baltimore, MD: Lippincott Williams & Wilkins; 2005.