Pharmacokinetic and Adsorptive Analyses of Administration of Oral Voriconazole Suspension via Enteral Feeding Tube in Intensive Care Unit Patients

Ryota Tanaka; Daiki Eto; Koji Goto; Yoshifumi Ohchi; Norihisa Yasuda; Yosuke Suzuki; Ryosuke Tatsuta; Takaaki Kitano; Hiroki Itoh

doi:10.1248/bpb.b20-00796

Abstract

For intensive care unit (ICU) patients, injectable voriconazole (VRCZ) is difficult to use because the patients often develop acute kidney injury. Since many ICU patients have consciousness disturbance, oral ingestion of tablet formulation is also difficult, and administration of a suspension via enteral feeding tube is required when using VRCZ. In this study, we investigated the in vitro adsorption property of oral VRCZ to feeding tube and performed pharmacokinetic analysis of VRCZ prepared by powdering and simple suspension for ICU patients. VRCZ was tube-administered to five ICU patients at a loading dose of 300 mg and plasma VRCZ concentrations before and at 1, 2, 4, 8, 12 h after the first dose were measured using HPLC. Pharmacokinetic parameters were calculated by non-compartmental model analysis. The recovery rate of VRCZ after infusion of the suspension through feeding tube was 89.8 ± 8.3%, but the cumulative rates after the first and second re-infusion were 102.7 ± 20.7 and 99.3 ± 10.3%, respectively, suggesting almost no residual drug in the tube after re-infusion. Metabolic phenotype was extensive metabolizer (EM) in two patients and intermediate metabolizer (IM) in three patients. The values of total clearance (CL_tot/F) calculated by moment analysis were 0.51 and 0.55 L/h/kg in two EM patients, and 0.09, 0.29 and 0.31 L/h/kg in three IM patients. The CL_tot/F was apparently lower in IM patients compared to EM. In conclusion, powdered and suspended VRCZ administered via enteral feeding tube showed pharmacokinetics depending on CYP2C19 gene polymorphism, similar to that observed in usual oral administration.

INTRODUCTION

Many patients in intensive care unit (ICU) have high-risk factors for the development of invasive candidiasis, such as central venous catheterization, frequent use of broad-spectrum antimicrobials, increased use of immunosuppressive agents or corticosteroids, and respiratory management with a ventilator. Empirical therapy for invasive candidiasis is recommended for patients with high fever and inflammatory findings persisting for a long time even after with administration of therapeutic antibacterial agents. The first choices of empiric anti-Candida therapy are echinocandin and fosfluconazole, while voriconazole (VRCZ) is one of the alternative drugs.

VRCZ is a second-generation triazole antifungal agent and is indicated for the treatment of invasive fungal infections by non-albicans Candida spp., Aspergillus spp. and Cryptococcus spp.^1,2) This drug has pharmacokinetic properties such as superior distribution to tissues and high bioavailability. The availability of different dosage forms including intravenous and oral formulations allows selection of the administration route. VRCZ is extensively metabolized in the liver, predominantly mediated by the CYP2C19 enzyme.^3,4) Allelic polymorphism of CYP2C19 has been reported and the metabolic phenotypes are categorized into extensive metabolizer (EM), intermediate metabolizer (IM) and poor metabolizer (PM) according to a combination of mutations of CYP2C19*2 (681G > A; rs4244285) and CYP2C19*3 (636G > A; rs17878459). The total clearance of VRCZ has been reported to be approximately 0.46 times for IM and 0.35 times for PM compared with EM.⁵⁾

Since VRCZ is water insoluble, the injection formulation contains sulfobutylether-β-cyclodextrin (SBECD) for solubilizing the drug. The urinary excretion rate of unchanged VRCZ is extremely low, but SBECD is not metabolized and is excreted in urine through the kidney. SBECD was reported to accumulate in patients with renal impairment.^6,7) Therefore, restricted use of intravenous VRCZ is recommended for patients with creatinine clearance of 50 mL/min or lower. As many ICU patients have renal impairment, use of intravenous VRCZ tends to be avoided. Although oral VRCZ can be used in ICU patients, many patients have difficulties ingesting the tablet by mouth due to disturbance of consciousness. Hence, it is necessary to powder and suspend the oral formulation and administer the suspension via the enteral feeding tube when oral VRCZ is used for ICU patients.

In this study, we investigated the in vitro adsorption property of VRCZ suspension to enteral feeding tube and performed pharmacokinetic analysis of the powdered and suspended VRCZ in ICU patients.

PATIENTS AND METHODS

Subject

The study enrolled five patients who received a powdered and suspended preparation of oral VRCZ administered via an enteral feeding tube between September 2016 and November 2017 in ICU of Oita University Hospital. The clinical and laboratory data were extracted from medical records. Creatinine clearance (CCr) was calculated using the Cockcroft–Gault equation.⁸⁾ This study was approved by the ethics committee (Approval No. 1025) and human genome research ethics committee of Oita University (Approval No. P-13-14). Written informed consent was obtained from either the patients or their legally authorized representatives.

Recovery Rate of VRCZ after Infusion through Enteral Feeding Tube

Recovery of VRCZ after infusion in enteral feeding tube was evaluated by an in vitro experiment. Using the simple suspension method,^9,10) VRCZ tablets (300 mg) were powdered using porcelain mortar and pestle and then suspended in 20 mL of distilled water at 55 °C. After stirring for 10 min, the suspension was drawn into an injection syringe and injected into a tube for enteral feeding (in our ICU, a balloon catheter for gastrointestinal imaging (Fuji Systems, Tokyo, Japan) is used). The enteral tube is derived from silicone rubber, and the surface is coated with a hydrophilic polymer. The three-way stopcock at the inlet is composed of polycarbonate and polypropylene. Then, the infused suspension was collected into a beaker. The enteral feeding tube was rinsed twice by infusing with 20 mL of distilled water at room temperature to ensure that no drug remains in the tube (a procedure routinely used clinically in our hospital). The first and second re-infused eluents were also collected. The VRCZ concentrations in each solution before and after infusion via the tube were measured by HPLC. The recovery rate after infusion through the tube was calculated using the concentrations before and after infusion.

Administration to Patients and Sample Collection

Using the simple suspension method,^9,10) VRCZ tablets were powdered and dissolved in 20 mL of distilled water at approximately 55 °C followed by incubation for 10 min. After cooling to around body temperature, the suspension was administered at a dose of 300 mg (loading dose) through the enteral feeding tube between meals, and flushed twice with 20 mL of distilled water at room temperature. Blood samples were collected before and at 1, 2, 4, 8, and 12 h after the first dose. Blood was collected from an indwelling arterial line into heparinized tubes. Blood samples were centrifuged at 1900 × g at 4 °C for 5 min, and plasma samples were frozen immediately and stored at −40 °C until measurement.

VRCZ Assay

VRCZ concentrations in plasma samples or suspensions were quantified using the partially modified method of Suzuki et al.¹¹⁾ To 500 µL of plasma or suspension in a glass tube, 30 µL of ketoconazole (500 µg/mL) in 100% methanol was added as internal standard, followed by 200 µL of sodium hydroxide (0.1 mol/L). The mixture was then briefly vortexed and extracted with 3 mL of diethyl ether for 5 min, followed by centrifugation at 1500 × g for 5 min. The organic layer was transferred into a glass tube and evaporated to dryness under a gentle stream of nitrogen at 40 °C, and the residue was reconstituted in 250 µL of mobile phase. Fifty microliters of the sample were injected into a HPLC system (Alliance model e2695; Waters, Milford, MA, U.S.A.) equipped with a reversed-phase C18 packed column (Cosmosil 5C18, 4.6 mm I.D. × 150 mm, Nacalai Tesque, Kyoto, Japan) followed by UV detection at 260 nm. Phosphate buffer (0.05 mol/L, pH 6.0)/acetonitrile/methanol (350/450/200, vol/vol/vol) was used as the mobile phase, at a flow rate of 0.7 mL/min. The column temperature was maintained at 40 °C throughout all experiments, while the sample temperature was maintained at 20 °C.

Genotyping Procedures for CYP2C19

A venous blood sample (5 mL) was obtained from each patient. Using 400 µL of blood, total DNA was prepared using the Maxwell^® 16 DNA Purification Kit (Promega, Tokyo, Japan). To determine the CYP2C19 mutant alleles, all samples were analyzed for the single nucleotide polymorphism 681G > A; rs4244285 (CYP2C19*2) or 636G > A; rs17878459 (CYP2C19*3). Allelic discrimination reaction was performed using TaqMan genotyping assays (C_26201809_30) in a LightCycler^® Nano System (Roche Applied Science, Penzberg, Germany). Subjects were classified into three metabolic phenotype subgroups: EM (CYP2C19*1/*1), IM (CYP2C19*1/*2 or CYP2C19*1/*3) and PM (CYP2C19*2/*2, CYP2C19*2/*3 or CYP2C19*3/*3).

Pharmacokinetic Analysis

Pharmacokinetic parameters were calculated by non-compartmental model analysis using the Excel^® (Microsoft) macro program MOMENT, created and validated by Tabata et al.^12,13) Log-linear trapezoidal rule was adopted in this calculation. To minimize Akaike’s information criterion, which was incorporated in the macro program, the last three points of the plasma concentration–time curve were used for linear regression. The area under the curve (AUC) was calculated with linear regression of integration to infinite time.

RESULTS

Recovery Rate of VRCZ in Feeding Tube Infusion

As shown in Table 1, the recovery rate after infusion of the VRCZ suspension was 89.8 ± 8.3%. The tube was rinsed twice by re-infusing distilled water to ensure no residual drug in the tube. The cumulative recovery rate after the first re-infusion was 102.7 ± 20.7% and that after the second re-infusion was 99.3 ± 10.3%. These results suggested that almost all of the powdered and suspended VRCZ administered via the enteral feeding tube would gain access to the gastrointestinal tract.

Table 1. Recovery Rates after Infusion of Voriconazole Suspension through an Enteral Feeding Tube

After infusion of suspension	After first re-infusion with water	After second re-infusion with water
89.8 ± 8.3%	102.7 ± 20.7%	99.3 ± 10.3%

Data are expressed as mean ± standard deviation (S.D.) (n = 7).

Plasma Concentration Profile and Pharmacokinetic Parameters of VRCZ in Each Patient

Patient characteristics are shown in Table 2. VRCZ was administered at a loading dose of 300 mg twice daily in all patients and the maintenance dose was 150 or 200 mg twice daily. Figure 1 shows the plasma concentration–time profiles after first loading of VRCZ in each patient. Plasma concentrations were confirmed in all patients, suggesting absorption of the drug after feeding tube administration of powdered and suspended VRCZ. Pharmacokinetic parameters of individual patients are shown in Table 3. The mean pharmacokinetic profile was: plasma concentration at 1 h after administration (C_max); 1.81 ± 0.64 µg/mL, AUC; 27.1 ± 27.1 h*µg/mL, t_1/2; 12.7 ± 8.1 h, mean residence time; 18.2 ± 11.9 h, total clearance/bioavailability (CL_tot/F); 0.35 ± 0.19 L/h/kg and distribution volume at steady state (Vd_ss)/F; 4.9 ± 1.3 L/h. Metabolic phenotype was EM in two patients and IM in three patients. The values of CL_tot/F calculated by moment analysis were 0.51 and 0.55 L/h/kg in two EM patients, and 0.09, 0.29 and 0.31 L/h/kg in three IM patients. The CL_tot/F was apparently lower in IM patients than in EM, which was similar to that observed in usual oral administration.⁵⁾

Table 2. Demographic and Clinical Data of Subjects

No. of patients (n)	5
Nos. of males/females (n)	4/1
Age (years)	65 [62–85]
Height (cm)	161.5 [150.0–167.6]
Body weight (kg)	53.9 [38.3–83.8]
Loading dose (mg/d)	300 × 2
Maintenance dose (mg/d)	150 × 2 or 200 × 2
AST (U/L)	39.4 [10.9–134.5]
ALT (U/L)	27.1 [3.0–148.1]
γ-GTP (U/L)	99.6 [13.3–648.1]
Albumin (g/dL)	2.89 [2.77–3.02]
Total bilirubin (mg/dL)	0.93 [0.53–3.84]
CRP (mg/dL)	6.87 [2.37–14.14]
CCr (mL/min)	33.3 [22.6–109.3]

Data are expressed as number of patients or median [range]. AST: aspartate aminotransferase, ALT: alanine aminotransferase, γ-GTP: γ-glutamyl transpeptidase, CRP: C-reactive protein, CCr: creatinine clearance.

Fig. 1. Individual Plasma Concentration–Time Profiles of Voriconazole (VRCZ) after the First Loading in Five Patients

Table 3. Metabolic Phenotype and Pharmacokinetic Parameters of Each Patient

	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
Metabolic phenotype	EM	IM	IM	EM	IM
Gender (f/m)	f	m	m	m	m
Age (year)	62	85	65	63	77
Weight (kg)	38.3	45.8	53.9	83.8	50.0
Dose/weight (mg/kg)	7.83	6.55	5.57	3.58	6.00
AST (U/L)	23.0	51.7	10.9	134.5	39.4
ALT (U/L)	11.9	81.1	3.0	148.1	27.1
γ-GTP (U/L)	13.3	99.6	61.2	648.1	105.0
Albumin (g/dL)	2.89	2.77	2.88	3.02	2.83
Total bilirubin (mg/dL)	0.91	1.09	0.93	3.84	0.53
CRP (mg/dL)	2.37	8.93	14.14	6.87	3.79
C_max (µg/mL)	1.26	2.40	2.07	0.99	2.32
AUC (h*µg/mL)	15.3	74.7	19.3	6.5	19.5
t_1/2 (h)	8.5	26.9	8.6	7.8	11.5
MRT (h)	12.7	39.1	12.7	10.7	15.7
CL_tot/F (L/h/kg)	0.51	0.09	0.29	0.55	0.31
Vd_ss/F (L/kg)	6.5	3.4	3.7	5.9	4.8
Primary disease	Apical ballooning syndrome Myasthenia gravis	Bladder cancer Postoperative aspiration pneumonia	Ischemic cardiomyopathy Mitral regurgitation	Aortic arch aneurysm	Left cervical abscess
Cause for intensive care unit management	Acute respiratory failure	Acute respiratory failure	Acute kidney injury Acute respiratory failure	Respiratory and circulatory management after aortic arch replacement	Sepsis

EM: extensive metabolizer, IM: intermediate metabolizer, f: female, m: male, AST: aspartate aminotransferase, ALT: alanine aminotransferase, γ-GTP: γ-glutamyl transpeptidase, CRP: C-reactive protein, C_max: plasma concentration at 1 h after administration, AUC: area under the curve, MRT: mean residence time, CL_tot: total clearance, F: bioavailability, Vd_ss: distribution volume at steady state.

DISCUSSION

Injectable VRCZ is difficult to use for ICU patients because they often have acute renal injury. On the other hand, since many ICU patients have consciousness disturbance, they have difficulties in ingesting tablets orally, and administration of a powdered and suspended preparation via the enteral feeding tube is required when using VRCZ. In this study, evaluation of adsorption of VRCZ to feeding tube confirmed almost 100% recovery rate after re-infusion with water twice. Furthermore, pharmacokinetic analysis of tube administration of a powdered and suspended VRCZ preparation revealed a similar plasma concentration profile as that of usual oral administration.⁵⁾

In our in vitro study, the recovery rate of VRCZ after the initial tube infusion was 89.8 ± 8.3%, suggesting that a part of the drug was not recovered from the tube (Table 1). VRCZ is highly lipophilic. When dissolved in solvents containing high proportions of water, some drugs, particularly highly lipophilic and peptide drugs, are known to adsorb to polypropylene tubes. According to our previous report, daptomycin, the cyclic lipopeptide drug, adsorbed to polypropylene tubes when water is selected as the solvent for the drug.¹⁴⁾ Moreover, since itraconazole, a triazole antifungal drug with high lipophilicity like voriconazole, also adsorbed to polypropylene tube in a solvent with a high proportion of water, 100% methanol was used as the solvent for the drug to avoid adsorption in our previous report.¹⁵⁾ As the surface of the enteral feeding tube is coated with a hydrophilic polymer, VRCZ is unlikely to show adsorptivity to the tube itself. In contrast, the three-way stopcock at the inlet is composed of polycarbonate and polypropylene. Therefore, adsorption of VRCZ to the three-way stopcock was suspected to be the reason why a part of the drug was not be recovered from the tube. However, the cumulative recovery rate after re-infusion with water was 102.7 ± 20.7%, confirming no adsorption to the tube after rinsing. We speculate that during the first infusion, some solid particles of the drug remained on the tube surface, but re-infusion with 20 mL of distilled water at room temperature allowed recovery of almost all the drugs. Although we did not evaluate prolonged stability at 55 °C, the recovery study suggested that almost all the VRCZ in the preparation would be administered via the feeding tube into the patient. However, recovery rates showed a high standard deviation, and that after the first re-infusion was ±20.7%. To investigate the in vitro recovery rate of in vivo infusion method, we used the quite concentrated VRCZ suspension of approximately 2500 µg/mL. The high standard deviation may stem from the insufficient dissolution owing to the insolubility of VRCZ in water and error by thousandfold diluting so that concentrations of the suspension fit within the calibration curve range of the quantification method. This is a limitation of our study.

Next, we evaluated the plasma concentration–time profiles and calculated pharmacokinetic parameters after the first (loading) dose to check whether VRCZ was absorbed following passage through the feeding tube. The values of CL_tot/F calculated by moment analysis in two EM patients were 0.51 and 0.55 L/h/kg, and those in three IM patients were 0.09, 0.29, and 0.31 L/h/kg. Similar to oral administration of VRCZ,⁵⁾ the clearance was lower than in IM phenotype than in EM phenotype. The calculated C_max per dose were 4.2 and 3.3 µg/mL/g in EM patients, and 8.0, 6.9, and 7.7 µg/mL/g in IM patients. Previous study indicated that mean maximum concentration per dose after an oral 400 mg dose of VRCZ was 7.2 µg/mL/g in EM phenotype and 8.2 µg/mL/g in IM phenotype.⁵⁾ The maximum concentration in EM phenotype of this study was slightly low, but there was almost no difference in the AUC per dose (this study: 51.2, 21.7, previous study: 33.9 h*µg/mL/g⁵⁾). The CL_tot/F of patient 2 was markedly lower than that of the other IM patients. As shown in Table 3, he was the oldest and had reduced hepatic reserve. Moreover, he suffered a mild hepatic injury, and his inflammatory response was increasing (Patient 3 had a higher CRP level, but the inflammatory response was in the process of declining). According to several previous reports, the pharmacokinetics of VRCZ is influenced by age,¹⁶⁾ hepatic function¹⁷⁾ and inflammation.¹⁸⁾ Hence, it is considered that these factors caused the lower CL_tot/F of patient 2.

This study is the first report on the adsorption to feeding tube and pharmacokinetics when powdered and suspended VRCZ was tube-administered. The results of in vitro study demonstrate that VRCZ did not adsorb to feeding tube and almost all the drug was recovered by re-infusion with distilled water twice. Furthermore, pharmacokinetic study in ICU patients revealed that the pharmacokinetics of feeding tube-administered powdered and suspended VRCZ was dependent on CYP2C19 gene polymorphism similar to that of usual oral administration. The limitation of this study is a very small sample of five patients. In the future, it is necessary to collect more cases and perform detailed pharmacokinetic analysis.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

1) Denning DW, Ribaud P, Milpied N, Caillot D, Herbrecht R, Thiel E, Haas A, Ruhnke M, Lode H. Efficacy and safety of voriconazole in the treatment of acute invasive aspergillosis. Clin. Infect. Dis., 34, 563–571 (2002).
2) Marks DI, Liu Q, Slavin M. Voriconazole for prophylaxis of invasive fungal infections after allogeneic hematopoietic stem cell transplantation. Expert Rev. Anti Infect. Ther., 15, 493–502 (2017).
3) Tan K, Brayshaw N, Tomaszewski K, Troke P, Wood N. Investigation of the potential relationships between plasma voriconazole concentrations and visual adverse events or liver function test abnormalities. J. Clin. Pharmacol., 46, 235–243 (2006).
4) Luong ML, Al-Dabbagh M, Groll AH, Racil Z, Nannya Y, Mitsani D, Husain S. Utility of voriconazole therapeutic drug monitoring: a meta-analysis. J. Antimicrob. Chemother., 71, 1786–1799 (2016).
5) Scholz I, Oberwittler H, Riedel KD, Burhenne J, Weiss J, Haefeli WE, Mikus G. Pharmacokinetics, metabolism and bioavailability of the triazole antifungal agent voriconazole in relation to CYP2C19 genotype. Br. J. Clin. Pharmacol., 68, 906–915 (2009).
6) Luke DR, Tomaszewski K, Damle B, Schlamm HT. Review of the basic and clinical pharmacology of sulfobutylether-beta-cyclodextrin (SBECD). J. Pharm. Sci., 99, 3291–3301 (2010).
7) Turner RB, Martello JL, Malhotra A. A worsening renal function in patients with baseline renal impairment treated with intravenous voriconazole: a systematic review. Int. J. Antimicrob. Agents, 46, 362–366 (2015).
8) Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron, 16, 31–41 (1976).
9) Kurata N, Komatsu C, Heito A, Mori Y. Examination and list of solid preparations being able to administer through feeding tubes. Jpn. J. Pharm. Health Care Sci., 27, 461–472 (2001).
10) Morita TO, Yamaguchi A, Kimura S, Fujii H, Endo K, Izumi K, Saito S, Minami H. Stability of lenalidomide suspension after preparation by a simple suspension method for enteral tube administration. J. Oncol. Pharm. Pract., 22, 579–583 (2016).
11) Suzuki Y, Tokimatsu I, Sato Y, Kawasaki K, Sato Y, Goto T, Hashinaga K, Itoh H, Hiramatsu K, Kadota J. Association of sustained high plasma trough concentration of voriconazole with the incidence of hepatotoxicity. Clin. Chim. Acta, 424, 119–122 (2013).
12) Tabata K, Yamaoka K, Kaibara A, Suzuki S, Terakawa M, Hata T. Moment analysis program available on Microsoft Excel^®. Drug Metab. Pharmacokinet., 14, 286–293 (1999).
13) Tabata K, Yamaoka K, Yasui H, Fukuyama T, Nakagawa T. Influence of pentobarbitone on in vivo local disposition of diclofenac in rat liver. J. Pharm. Pharmacol., 48, 866–869 (1996).
14) Tanaka R, Suzuki Y, Goto K, Yasuda N, Koga H, Kai S, Ohchi Y, Sato Y, Kitano T, Itoh H. Development and validation of sensitive and selective quantification of total and free daptomycin in human plasma using ultra-performance liquid chromatography coupled to tandem mass spectrometry. J. Pharm. Biomed. Anal., 165, 56–64 (2019).
15) Suzuki Y, Tanaka R, Oyama N, Nonoshita K, Hashinaga K, Umeki K, Sato Y, Hiramatsu K, Kadota JI, Itoh H. Sensitive and selective quantification of total and free itraconazole and hydroxyitraconazole in human plasma using ultra-performance liquid chromatography coupled to tandem mass spectrometry. Clin. Biochem., 50, 1228–1236 (2017).
16) Liu Y, Qiu T, Liu Y, Wang J, Hu K, Bao F, Zhang C. Model-based voriconazole dose optimization in chinese adult patients with hematologic malignancies. Clin. Ther., 41, 1151–1163 (2019).
17) Lin XB, Huang F, Tong L, Xia YZ, Wu JJ, Li J, Hu XG, Liang T, Liu XM, Zhong GP, Cai CJ, Chen X. Pharmacokinetics of intravenous voriconazole in patients with liver dysfunction: A prospective study in the intensive care unit. Int. J. Infect. Dis., 93, 345–352 (2020).
18) Veringa A, Ter Avest M, Span LF, van den Heuvel ER, Touw DJ, Zijlstra JG, Kosterink JG, van der Werf TS, Alffenaar JC. Voriconazole metabolism is influenced by severe inflammation: a prospective study. J. Antimicrob. Chemother., 72, 261–267 (2017).

Corresponding author

Register with J-STAGE for free!