Pharmacokinetics and Dialytic Clearance of Isavuconazole during In Vitro and In Vivo Continuous Renal Replacement Therapy

The pharmacokinetics (PK) and dialytic clearance of isavuconazole in vitro and in 7 solid-organ transplant patients undergoing continuous renal replacement therapy (CRRT) were evaluated.

ABSTRACT

The pharmacokinetics (PK) and dialytic clearance of isavuconazole in vitro and in 7 solid-organ transplant patients undergoing continuous renal replacement therapy (CRRT) were evaluated. In vivo, the mean (± standard deviation [SD]) plasma PK parameters of isavuconazole were as follows: maximum concentration of drug in serum (C_max), 4.00 ± 1.45 mg/liter; minimum concentration of drug in serum (C_min), 1.76 ± 0.76 mg/liter; half-life (t_1/2), 48.36 ± 29.78 h; volume of distribution at steady state (V_ss), 288.78 ± 182.11 liters, clearance at steady state (CL_ss), 4.85 ± 3.79 liters/h; and area under the concentration-time curve (AUC), 54.01 ± 20.98 mg · h/liter. Transmembrane clearance represented just 0.7% of the total isavuconazole clearance. These data suggest that isavuconazole is not readily removed by CRRT and no dose adjustments are necessary.

INTRODUCTION

Invasive fungal infections (IFIs) are a significant cause of morbidity, mortality, and excess health care costs (1–4). Guidelines recommend prophylaxis when there is increased risk of invasive aspergillosis. Therefore, mold-active triazoles are frequently used for prophylaxis and treatment of IFIs in bone marrow and solid-organ transplant populations (5–7).

Isavuconazole is a broad-spectrum mold-active triazole approved for the treatment of invasive aspergillosis and mucormycosis infections in adults (8). Isavuconazole possesses important advantages over other mold-active triazoles, including high bioavailability, linear and dose-proportional pharmacokinetics (PK), once daily administration, and improved solubility (9). Together, these factors along with a more favorable drug-drug interaction and toxicity profile compared to those of other mold-active triazoles have led to the increasing use of isavuconazole for prophylaxis and treatment of IFIs (6, 10–12).

Patients who require prophylaxis or treatment with mold-active triazoles frequently suffer from pathophysiological derangements such as acute kidney injury (AKI), often necessitating renal replacement therapies, including continuous renal replacement therapy (CRRT). Robust PK data in patients receiving CRRT are scarce and often include only small numbers of critically ill patients on many different forms of CRRT with heterogeneous flow rates, filter types, dosing, and sampling schemes, making it difficult to draw meaningful conclusions (13–17). As such, in vitro CRRT models are useful for generating precise assessments of sieving/saturation coefficients across different modes, flow rates, filter types, and points of dilution while eliminating the variability introduced by the patient. These models can be used to guide dosing in the absence of, or when combined with, in vivo data (18). These in vitro models may be especially valuable for agents that have pharmacologic properties not intuitively predictive of significant removal by CRRT given the multitude of underappreciated external factors other than the drug’s properties that contribute significantly to removal by CRRT, such as effluent/dialysate flow rate, filter charge and adsorption, and point of replacement fluid dilution. This has been highlighted by study of the lipoglycopeptide dalbavancin, which has a large molecular size, high protein binding, and minimal renal elimination yet demonstrated a 50% reduction in half-life from 8 days to 4 days at in vitro CRRT flow rates of 1 to 6 liters/h, respectively (19). Currently there are no data to inform the PK or support dosing of isavuconazole during CRRT. As such, the objective of this study was to evaluate the PK and dialytic clearance of isavuconazole during in vitro CRRT and in solid-organ transplant patients receiving either continuous veno-venous hemofiltration (CVVH) or continuous veno-venous hemodiafiltration (CVVHDF).

(Results of this study were presented in part at the 29^th European Congress of Clinical Microbiology & Infectious Diseases in Amsterdam, Netherlands, as abstract 2885 [20].)

In vitro CRRT was simulated using a Prismaflex 7.2 system (Baxter Healthcare Corporation, Deerfield, IL) with a fresh 0.9-m² acrylonitrate copolymer (AN69) filter in three different modes: CVVH, continuous veno-venous hemodialysis (CVVHD), and CVVHDF. Whole bovine blood (Densco Marketing Inc., Woodstock, IL) anticoagulated with 20 units/ml of heparin was used as the vehicle for all experiments. One liter of blood was warmed to 37°C, continuously stirred, and circulated for at least 5 min prior to each experiment to allow adequate exposure of the filter to blood proteins. The Prismaflex circuit was primed according to the manufacturer’s instructions and the blood flow and total dialysate/replacement fluid flow rates were fixed at 200 ml/min and 2 liters/h, respectively. PrismaSOL and PrismaSATE (Baxter Healthcare Corporation, Deerfield, IL) were used as replacement fluid (CVVH and CVVHDF) and dialysate (CVVHD and CVVHDF), respectively, and added 50% pre-/50% postfilter during CVVH and 100% postfilter during CVVHDF at a rate of 0.25 liters/h. Isavuconazonium sulfate (Cresemba; Astellas Pharma, Inc.) was reconstituted per the manufacturer’s instructions, added to the blood reservoir, and allowed to equilibrate for 1 min prior to each experiment. Serial undiluted prefilter blood samples were collected in K₂ EDTA tubes prior to and at 10, 20, 30, and 60 min following drug equilibration. Postfilter blood and effluent samples were collected at 10 and 30 min. Blood samples were centrifuged at 1,500 × g for 10 min, and the resultant supernatant plasma and ultrafiltrate samples were frozen at −80°C within 30 min of collection until analysis.

The data from a subset of six solid-organ transplant patients (SOT) receiving CVVHDF from a previously published PK study in 26 SOT patients (6) from the University of Pittsburgh Medical Center (UPMC) were obtained and analyzed along with data from one patient from the University of Illinois Hospital and Health Sciences System (UIHHSS) receiving CVVH. Information regarding machine and filter types along with blood and dialysate/effluent flow rates utilized during CRRT are displayed in Table 1. No anticoagulation was added to the CRRT circuit for any patient at either study site. All patients received isavuconazonium sulfate 372 mg (equivalent to 200 mg isavuconazole) intravenous (i.v.) every 8 h infused over 1 h for six doses, followed by 372 mg i.v. once daily. Following a minimum of seven doses, serial prefilter blood samples were collected prior to (0 h) and at 1, 2, 4, 6, 8, 12, 16 (UPMC only), and 24 h postdose. Simultaneous postfilter and effluent samples were collected from the UIHHSS patient at 1, 8 (effluent only), 12, and 24 h. Samples were centrifuged at 1,500 × g for 10 min, and the resultant supernatant plasma and ultrafiltrate samples were frozen at −80°C within 30 min of collection until analysis. Plasma samples obtained from prefilter blood samples 10 min after drug equilibration from each in vitro and from the UIHHSS patient at 1, 12, and 24 h following administration were subsequently centrifuged using a Centrifree ultrafiltration device (Merck Millipore Ltd., Carrigtwohill, Ireland) on a fixed-angle rotor at 2,000 × g for 30 min, with the resulting bound and unbound samples frozen at −80°C within 30 min until analysis.

TABLE 1.

Baseline demographics and CRRT parameters of solid-organ transplant patients receiving i.v. isavuconazole

Patient	Sex	Age (yrs)	ht (cm)	wt (kg)	BMIa (kg/m2)	Solid-organ transplant	No. of doses prior to samplingb	CRRT machine	CRRT mode	Filter type	Blood flow rate (ml/min)	Effluent flow rate (liters/h)
1	M	67	175	122	39.8	Kidney	8	NxStage	CVVH	PSc	200	2.0d
2	M	59	168	69	24.4	Lung	15	Prismaflex	CVVHDF	AN69	250	2.1e
3	F	33	145	41	19.5	Lung	12	Prismaflex	CVVHDF	AN69	300	1.5e
4	M	66	175	57	18.6	Lung	7	Prismaflex	CVVHDF	AN69	250	2.5e
5	M	60	175	96	31.3	Lung	18	Prismaflex	CVVHDF	AN69	200	2.5e
6	F	63	158	42	16.8	Lung	13	Prismaflex	CVVHDF	AN69	300	2.5e
7	M	39	183	89	26.6	Heart	11	Prismaflex	CVVHDF	AN69	250	2.3e

^aBMI, body mass index.

^bAll patients received the equivalent of isavuconazole 200 mg i.v. every 8 h for six doses followed by 200 mg i.v. once daily.

^cPS, polyethersulfone.

^dPrismaSOL added 100% prefilter.

^ePrismaSOL added postfilter at a rate of 0.25 liters/h.

Isavuconazole concentrations obtained from in vitro CRRT experiments and the UIHHSS patient were quantified using liquid chromatography tandem mass-spectrometry (Keystone Bioanalytical, North Wales, PA). The calibration range was 0.05 to 20 mg/liter with intra- and interassay error rates of ± 10%. Concentrations from UPMC patients were quantified as previously described (6).

Steady-state pharmacokinetic parameters for isavuconazole were estimated from observed prefilter plasma concentrations via noncompartmental analysis in Phoenix WinNonlin version 8.1 (Certara USA Inc., Princeton, NJ). As in vitro experiments were run for 1 h, the area under the concentration-time curve from 0 to last (AUC_0–last) was multiplied by 24 to show proportional AUC_0–24. Calculations for the estimation of removal of isavuconazole by CRRT, when appropriate, were as follows (21):

• •Extraction ratio (%): [(prefilter concentration − postfilter concentration)/prefilter concentration] × 100
• •Sieving coefficient (SC): ultrafiltrate concentration/prefilter plasma concentration
• •Saturation coefficient (SA): dialysate concentration/prefilter plasma concentration (22)
• •Transmembrane clearance (CL_TM):
  • ∘CVVH: SC × ultrafiltrate flow rate × correction factor for prefilter fluid replacement
  • ∘CVVHD: SA × dialysate flow rate
  • ∘CVVHDF: SA × total effluent flow rate × correction factor for prefilter fluid replacement
    • ▪Correction factor: blood flow rate/(blood flow rate + prefilter replacement fluid rate) (23)

Institutional review board approval was obtained at UPMC, and all patients provided written informed consent. For the UIHHSS patient, this work involved the use of a marketed drug in the course of routine medical practice and therefore did not meet the institutional criteria with respect to ethical guidelines for human subjects research.

Prefilter plasma pharmacokinetic parameters of isavuconazole during in vitro CRRT are summarized in Table 2. Overall, the pharmacokinetic parameters of isavuconazole were similar during each CRRT mode. For all in vitro experiments, the extraction ratio, SC, SA, and CL_TM of isavuconazole could not be determined, as concentrations in the effluent samples were below the limit of quantitation. Protein binding across the 3 in vitro CRRT modes tested ranged from 97.5% to 98.5%.

TABLE 2.

Plasma pharmacokinetic parameters of isavuconazole during in vitro CRRT

Therapy	Cmax (mg/liter)	Cmin (mg/liter)	AUC0–last (mg · h/liter)	AUC0–24 (mg · h/liter)	CLT (liters/h)	V (liters)
CVVH	1.39	0.47	0.58	13.92	3.85	4.63
CVVHD	1.68	0.46	0.63	15.12	4.3	3.73
CVVHDF	0.96	0.44	0.59	14.16	3.22	5.22

Baseline demographics and CRRT characteristics for the seven solid-organ transplant patients receiving i.v. isavuconazonium sulfate are summarized in Table 1. Prefilter plasma pharmacokinetic parameters of isavuconazole in the patient receiving CVVH and mean (± standard deviation [SD]) prefilter plasma pharmacokinetic parameters of patients receiving CVVHDF are also displayed in Table 3 along with parameters previously reported in healthy subjects and patients with invasive fungal diseases.

TABLE 3.

Plasma pharmacokinetic parameters of isavuconazole in solid-organ transplant patients receiving CRRT, healthy subjects, and patients with invasive fungal infection^a

Group	Cmax (mg/liter)	Cmin (mg/liter)	t1/2 (h)	AUC0–24 (mg · h/liter)	CLSS (liters/h)	VSS (liters)
CVVH (n = 1)	1.73	0.47	29.33	15.03	13.31	564.95
CVVHDF (n = 6)	4.38 ± 1.15	1.97 ± 0.54	51.53 ± 31.30	60.51 ± 13.18	3.44 ± 0.73	242.75 ± 148.33
Healthy subjects (n = 6)b	2.55 ± 0.88		117 ± 17.6c	33.6 ± 9.67	3.19 ± 0.90	542 ± 229c
IFI patients (n = 136)d				87.1 ± 41	2.5 ± 1.6	361.2 ± 166.3

^aData are presented as means ± SDs.

^bSubjects received a single i.v. loading dose equivalent to 200 mg isavuconazole followed by daily i.v. maintenance doses equivalent to 100 mg isavuconazole for 14 days. All doses were infused over 1 h (28).

^cn = 4.

^dPatients received i.v. or oral (p.o.) dose of isavuconazole equivalent to 200 mg every 8 h for 48 h followed by i.v. or p.o. isavuconazole 200 mg once daily (20).

The prefilter plasma pharmacokinetic parameters for the UIHHSS CVVH patient are shown in Table 3, and the plasma concentration-time profile is displayed in Fig. 1. The extraction ratios of isavuconazole at 12 and 24 h after the start of the infusion were 11.86% and 1.11%, respectively. The SC was only measurable 1 h after the start of the infusion and was 0.045, corresponding to a CL_CVVH of 0.09 liters/h, representing only 0.68% of the total clearance. Unbound prefilter plasma concentration of isavuconazole at hours 1, 12, and 24 postdose were all below the lower limit of quantification, resulting in a calculated plasma protein binding of isavuconazole of ≥97%.

FIG 1.

Prefilter plasma concentration-time profiles of isavuconazole in solid-organ transplant patients receiving either CVVH (dashed line, open circles) or CVVHDF (dashed line, open triangles). Mean values for CVVHDF patients are displayed with error bars representing standard deviations. The y axis is in the log scale.

Demographics of the six solid-organ transplant recipients initiated on CVVHDF while receiving i.v. isavuconazole for prophylaxis are displayed in Table 2, and mean (± SD) prefilter plasma pharmacokinetic parameters are shown in Table 3 with corresponding concentration-time profiles displayed in Fig. 1. Compared to that for the CVVH patient, patients receiving CVVHDF had a lower mean clearance at steady state (CL_SS; 13.31 versus 3.44 ± 0.73 liters/h, respectively) and volume of distribution at steady state (V_SS; 564.95 versus 242.75 ± 148.33 liters, respectively), resulting in higher AUC values (15.03 versus 60.51 ± 13.18 mg · h/liter, respectively).

This study comprehensively evaluated the pharmacokinetics and dialytic clearance of isavuconazole utilizing both in vitro and in vivo analyses, including multiple study sites and two different CRRT machines, modes, filter types, and flow rates. To our knowledge, these are the first data available to guide dosing of isavuconazole in this population and suggest that isavuconazole is not readily removed by CRRT; therefore, dosage adjustments are not necessary.

The degree of protein binding is a known factor affecting the clearance of antimicrobials during CRRT, and removal of typically highly protein bound antimicrobials may be increased in critically ill patients with hypoalbuminemia receiving CRRT (23–26). Previous studies suggest that plasma protein binding may also play an important role in the clearance of isavuconazole during dialysis (6). However, our results suggest that the protein binding of isavuconazole is largely unaffected by CRRT, as unbound concentrations were often unmeasurable, leading to a calculated protein binding of >97% in this study across all experiments, including in the critically ill UIHHSS patient despite severe hypoalbuminemia (baseline albumin, 1.6 g/dl).

The isavuconazole PK observed in the UIHHSS patient and during in vitro experiments differ from those reported for other populations in Table 3, including the patients included from Wu et al. (6). This is likely due to the large body habitus and critical illness secondary to the fatal IFI experienced by the UIHHSS patient in addition to differences in the machine type, CRRT mode, point of replacement fluid addition, filter type used in vivo and in vitro, and isavuconazole dosing and sampling schemes. Additionally, there may have been a decreased conversion rate of isavuconazonium sulfate to isavuconazole in vitro due to a relatively lower content of butylcholinesterase in bovine serum (27). Although routine therapeutic drug monitoring of isavuconazole is not currently recommended (28), the variability observed across these data may indicate a need for this in the future, especially in special patient populations.

Although as many different CRRT machines, filter types, filtration points, and flow rates as possible were included in this study, the results may not be representative of all modalities of CRRT. Additional limitations include the lack of evaluation of adsorption to the CRRT circuit and incomplete elimination phase sampling data; therefore, the reported half-lives in this study should be interpreted with caution. Finally, protein binding data were only obtained in one patient and therefore warrant further confirmation.

The data obtained herein will help to guide dosing of isavuconazole in this vulnerable patient population suffering from significant pathophysiologic and PK alterations. Our results suggest that no dosage adjustments are necessary in patients receiving isavuconazonium sulfate during CRRT, analogous to the package insert recommendation for patients with end-stage renal disease (ESRD) on hemodialysis (HD) (8). These data should be confirmed in future studies including more patients and employing additional CRRT settings.