Development and Validation of a Simple, Fast, and Sensitive LC/MS/MS Method for the Quantification of Oxfendazole in Human Plasma and Its Application to Clinical Pharmacokinetic Study

Abstract

The most popular standard treatments for soil transmitted helminths in humans including mebendazole, albendazole, levamisole, and pyrantel pamoate, show greatly variable efficacy against different species of parasites and have unfavorable pharmacokinetic characteristics, such as short half-life. The transition of oxfendazole, a potent broad-spectrum anthelmintic with long half-life, from veterinary medicine to human use has been considered as a promising approach. However, analytical methods for the quantitative detection of oxfendazole in human matrix are very limited and lack sensitivity. In this study, we have developed a high-performance liquid chromatography-tandem mass spectrometry (LC/MS/MS) method for the quantification of oxfendazole in human plasma using albendazole as an internal standard. The established method was fully validated with lower limit of quantitation (LLOQ) of 0.5 ng/mL and linearity in the range of 0.5 –1000 ng/mL; intra-day and inter-day accuracies ranged from 2.6 to 9.5% for 3 quality control levels (1.5 ng/mL, 75 ng/mL, and 750 ng/mL) and LLOQ; intra-day and inter-day precision was ≤13.6% for quality controls and ≤15.1% for LLOQ; matrix factor and extraction recovery were consistent with coefficient of variation of less than 15.0%. Other parameters including matrix selectivity, injection carryover, reinjection reproducibility, hemolysis effect, interference of analyte with internal standard, dilution integrity, freeze/thaw stability, whole blood stability, and stock solution stability were also validated and met the acceptance criteria. The assay was successfully applied to quantify oxfendazole plasma concentration in healthy adult volunteers after the administration of multiple oral doses of oxfendazole.

1. Introduction

Soil transmitted helminthiasis is one of the most common parasitic diseases and has affected approximately 24% of the world’s population, especially in highly endemic areas [1]. Soil transmitted helminthiasis is mainly caused by three different species of helminths, namely Ascaris lumbricoides (roundworm), hookworm and Trichuris trichiura (whipworm). Benzimidazoles are the most frequently used anthelmintics [2]. Among the benzimidazoles, mebendazole and albendazole are currently marketed for human use, while the other benzimidazoles are only used in veterinary medicine. Despite the extensive use of mebendazole and albendazole for soil transmitted helminth infections, they are not ideal due to their varied and/or poor efficacy profiles [1, 3] as well as unfavorable pharmacokinetics. For example, significant inter-individual variability in albendazole half-life, ranging from 2.8 to 9 hours, was observed in human following oral administration [2]. Because of the pressing clinical need for alternative anthelmintics, the possible transition of benzimidazoles from veterinary use to human use has been considered as an attractive strategy [4].

Oxfendazole is a potent, broad-spectrum anthelmintic which is currently marketed for treating lungworms and enteric helminths in food producing animals. Pharmacokinetic studies conducted in dogs and sheep shown that oxfendazole half-life was 2 times longer than albendazole half-life [5, 6] and, consequently, can maintain effective concentrations in both plasma and intestinaltract for a longer period of time. Considering its potentially favorable pharmacokinetic profile, oxfendazole represents an attractive anthelmintic candidate for transition to human use. However, no analytical method has currently been published to detect oxfendazole in human matrix. A number of high performance liquid chromatography (HPLC) methods have been established to measure oxfendazole concentrations in dairy [7], meat products [8], and animal blood samples [4, 9–14]. Most of these methods have sensitivity ranging from 5 to 50 ng/mL [4, 5, 9, 11, 13–15] and/or involve a relatively long sample extraction process [5, 7, 9–15]. Recently, an oxfendazole first-in-human (FIH) single escalating dose study has been carried out in healthy adult volunteers (ClinicalTrials.gov Identifier: NCT02234570) [16]. Coupled with this FIH clinical trial, an ultra high-performance liquid chromatography-mass spectrometry (UHPLC/MS) method has been developed to measure oxfendazole and its metabolites in both human plasma and human urine [16]. This represents the first bioanalytical method for oxfendazole quantification in human matrix. In that method, the lower limit of quantitation (LLOQ) of oxfendazole was 2 ng/mL and sample preparation involved solid phase extraction which is a labor-intensive procedure. At present, a multiple escalating dose clinical trial of oxfendazole in healthy adult volunteers is underway (ClinicalTrials.gov Identifier: NCT03035760) [17]. To better characterize oxfendazole pharmacokinetics, especially its half-life, in human, assay sensitivity is critical for capturing the terminal phase of oxfendazole disposition. The aim of this study is to develop and validate a simple and sensitive method for oxfendazole quantification in human plasma using liquid chromatography-tandem mass spectrometry (LC/MS/MS). Albendazole was chosen as the internal standard due to the high similarity between albendazole and oxfendazole in term of molecular structure, and the absence of albendazole in the expected biological samples. In our method, samples were extracted using the simple and fast protein precipitation technique, and oxfendazole can be accurately quantified to 0.5 ng/mL. Finally, with broad linear range (0.5 – 1000 ng/mL) and reliable performance after 15 folds of sample dilution, the validated assay can reliably quantify very low plasma oxfendazole concentration at the terminal phase and very high plasma oxfendazole peak concentration at steady-state at all dose levels tested.

2. Materials and Methods

2.1. Chemicals and Reagents

Oxfendazole (≥98%) and albendazole (≥98%) were acquired from Sigma-Aldrich Co (St. Louis, MO, USA). LC/MS grade formic acid (99.5+%), water, acetonitrile, and methanol were products of Fisher Scientific (Fairlawn, NJ, USA). Sodium heparin human plasma and 2% hemolyzed sodium heparin human plasma were obtained from BioreclamationIVT (Westbury, NY, USA).

2.2. Instrumentation and Analytical Conditions

2.2.1. Chromatography

Shimadzu UFLC LC-AD20 system (Shimadzu, Japan), which consists of a DGU–20A3 Degasser, a CTO–20A Column Oven, a SIL-20AC UFLC Autosampler, and a LC-20AD Binary Pump, was used to perform liquid chromatography. Ten μL of sample was injected onto a Phenomenex Synergi™ C18 column (150 × 2 mm, 4 μm) coupled with a Phenomenex 4 × 2 mm security guard cartridge (both from Phenomenex, Torrance, CA, USA). The column was maintained at 40°C. Oxfendazole and albendazole were separated under isocratic elution of water-acetonitrile (50:50, v/v); both mobile phases were conditioned with 0.1% formic acid. The total flow rate was 0.2 mL/min, with the total run time for each sample being 6.0 minutes. Autosampler was kept at 4°C. Ion suppression was minimized by diverting the flow from LC column to waste for the first 1.5 minutes using a Valco® diverter valve (VICI®, Houston, TX). Under these conditions, oxfendazole and albendazole typically eluted after 2.80 and 4.07 minutes, respectively.

2.2.2. Mass spectrometry

Oxfendazole and albendazole detection and quantification was performed using API4000 triple quadrupole mass spectrometer (AB Sciex LLC, Redwood City, CA, USA) with a TurboIonSpray probe in positive mode operating at 5500 V. To enhance sensitivity and specificity, multiple reaction monitoring (MRM) was applied. Fragmentation was induced with a collision energy (CE) of 30 eV. Curtain gas supply was set at 10 psi (approximately 70 kPa) and temperature was 350°C. Additional MS conditions included an entrance potential (EP) of 10 V, a declustering potential (DP) of 60 V, and a collision exit potential of 11 V. Data acquisition and processing was completed using Analyst software (version 1.6.2, Ab Sciex, Framingham, MA). The m/z transition used to monitor oxfendazole was 316.1 for the protonated molecule to 191.3 for the fragment ion. The m/z ratio of the parent ion and product ion for albendazole were 266.3 and 234.1, respectively.

2.3. Preparation of Calibration Standards and Quality Control Samples

Stock solutions of oxfendazole and albendazole at 100 μg/mL were prepared in methanol. The storage condition for the stock solutions was −20°C. Oxfendazole working solutions for calibration standards were prepared by serial dilution with methanol to reach final concentrations of 10000, 5000, 2500, 1000, 500, 100, 50, 10, and 5 ng/mL. Oxfendazole working solution at 7500, 750 and 15 ng/mL for quality control (QC) samples and albendazole working solution at 2500 ng/mL were prepared in the same way.

Oxfendazole calibration standards and QC samples were prepared by spiking 10 μL of the appropriate oxfendazole working solution and 10 μL of albendazole working solution in 90 μL of blank human plasma with sodium heparin, resulting in calibration standard concentrations of 1000, 500, 250, 100, 50, 10, 5, 1, and 0.5 ng/mL and QC concentrations of 750, 75, and 1.5 ng/mL. The final concentration of albendazole in oxfendazole calibration standards and QC samples was 250 ng/mL. For zero sample (i.e. without analyte but with IS), 10 μL of methanol was added instead of oxfendazole working solution. For blank sample (i.e. with no analyte and no IS), 20 μL of methanol was spiked in 90 μL of blank human plasma.

2.4. Sample Extraction

Following the preparation of calibration standards and QC samples, protein precipitation was initiated by adding 200 μL of acetonitrile to each sample. After mixing on a vortex mixer for 30 seconds, the samples were centrifuged at 18000 × g at 4°C for 20 min. Subsequently, 100 μL of the supernatant was transferred into LC/MS vials for LC/MS/MS analysis.

2.5. Method Validation

2.5.1. Sensitivity and linearity

Linearity within the standard concentration range, 0.5 – 1000 ng/mL, in human plasma was evaluated. The correlation between peak area ratio and oxfendazole nominal concentration is characterized by 1/x² weighted linear regression. The calibration curve was established if the accuracies of at least 75% calibration standards, including lower limit of quantitation (LLOQ) and upper limit of quantitation (ULOQ), fell within the ±15.0% range (±20.0% at LLOQ) and the correlation coefficient (r²) was >0.980. The lowest concentration on the calibration curve that could be determined with 80.0 – 120.0% accuracy and had a coefficient of variation (CV) not exceeding 20.0% was chosen as the LLOQ.

2.5.2. Selectivity

To evaluate matrix selectivity, aliquots from 6 different lots of blank human plasma was spiked with methanol (i.e. blank sample) or IS (i.e. zero sample), extracted, and analyzed. The acceptance criteria included: 1) peak area present on the chromatogram, from the zero samples, at the elution time of oxfendazole was no more 20.0% of oxfendazole response from LLOQ samples; and 2) for IS channel, the interference peak from the blank must not exceed 5.0% of IS signal from the zero sample.

2.5.3. Intra-day and inter-day precision and accuracy

Intra-day and inter-day accuracy and precision were investigated by analyzing QC samples prepared at four different concentrations (QC LLOQ, QC low, QC med, and QC high). Six replicates at each concentration level were analyzed on 3 different days. Following are the acceptance criteria: a) at each concentration, the overall accuracy must be within ±15.0%, except for the QC LLOQ that must be within ±20.0%; b) CV should be no more than 15.0% at each concentration level and 20.0% for QC LLOQ; and c) at least 50% of the samples at each QC level and at least 2/3 of all QC samples must meet criteria (a).

2.5.4. Injection carryover

Carryover was evaluated by injecting a blank sample subsequent to the injection of a ULOQ standard (1000 ng/mL for oxfendazole). The peak area of oxfendazole observed in the blank was compared with oxfendazole response from the LLOQ (i.e. 0.5 ng/mL); and albendazole (IS) peak area in the blank was compared with the mean albendazole peak area in accepted calibration standards and QC samples within that day. Carryover in the blank sample following ULOQ no greater than 20.0% of the LLOQ for oxfendazole, and 5.0% for IS was considered acceptable.

2.5.5. Reinjection reproducibility

After one of the intra-runs finished successfully, all QC samples (six replicates of high, medium and low QCs) and standards were kept at autosampler temperature (4°C) for at least 24 hrs. These QCs and standards were then re-injected to determine processed sample reinjection reproducibility. Acceptance criteria in “Linearity” and “Accuracy and Precision” also applied here.

2.5.6. Matrix effect

Six distinguished lots of human plasma were used for preparation of extracted blank matrix. One aliquot of each lot was extracted as described in section 2.4 without adding IS. Following extraction, appropriate amounts of oxfendazole and albendazole were added in all extracted blank samples so that the oxfendazole concentrations were at high and low QC levels.Neat oxfendazole solution (in triplicate) at QC high and QC low concentrations was fortified with IS and diluted using the same dilution factor as the matrix samples. The following equation was used for matrix factor calculation:

$$\text{Matrix factor}=\frac{\text{Peak area ratio in the presence of matrix}}{\text{Peak area ratio in the absence of matrix}}$$

Matrix effect was evaluated using the CV of the IS-normalized matrix factor from 6 lots of human plasma. The variability in IS-normalized matrix factors (CV) was considered acceptable if it was no more than 15.0% at each QC level.

2.5.7. Extraction recovery

Extraction recovery was assessed by comparing the IS-normalized oxfendazole peak area of the un-extracted samples with that of the extracted samples. For un-extracted samples, blank human plasma was fortified with oxfendazole and albendazole prior to extraction. Meanwhile, oxfendazole and IS working solution were added to supernatant recovered after blank plasma extraction. Three QC levels (750, 75 and 1.5 ng/mL) were evaluated in triplicate. The CV of the recovery for each QC level should be no more than 15.0%.

$$\text{Recovery}=\frac{\text{Peak area ratio in un-extracted samples}}{\text{Peak area ratio in extracted samples}}$$

It should be noted that, according the current experimental method, the calculated recovery reflects the relative extraction recovery of oxfendazole to that of albendazole from the biological matrix.

2.5.8. Hemolysis effect

To evaluate the effect of blood hemolysis, QC samples at 15 and 750 ng/mL were prepared in 2% hemolyzed plasma in six replicates, extracted, and analyzed with calibration standards preparedin non-hemolyzed plasma. Blood hemolysis effect was considered acceptable if, at each QC level, the accuracy was within ±15.0% and CV was no more than 15.0%.

2.5.9. Interference of oxfendazole with IS

Interference of analyte with IS was investigated by analyzing 3 replicates of ULOQ samples (i.e. 1000 ng/mL oxfendazole in human plasma) without the addition of albendazole. Interference was considered insignificant if mean albendazole peak area observed in ULOQ samples was ≤5.0% mean albendazole response in all accepted calibration standards and QC samples in the same batch.

2.5.10. Dilution integrity

To evaluate dilution integrity, oxfendazole samples with final concentration of 15000 ng/mL (15 times ULOQ) were prepared in six replicates and diluted with blank human plasma to reach the final concentration of 1000 ng/mL (15-fold dilution). Diluted samples were processed following the method described in section 2.4 and analyzed with calibration standards prepared on the same day. Accuracy and precision within ±15.0% were set as acceptance criteria.

2.5.11. Stability

a). Freeze (−80°C)/thaw stability

To evaluate oxfendazole stability in human plasma after 5 freeze/thaw cycles, six replicates of low and high QCs were prepared and stored at –80°C for at least 24 hours before thawing for the first time. In the following freeze/thaw cycles, samples were stored at –80°C for at least 12 hours. On the last freeze/thaw cycle, oxfendazole concentration in QC samples were measured based on freshly prepared calibration standards. Stability acceptance criteria were set as observed oxfendazole concentration in the range of 85.0 –115.0% of nominal concentration and CV not exceeding 15.0% at each QC level.

b). Whole blood stability

It is important to identify the period of time and condition at which whole blood collected from study participants can be stored before centrifugation without changing oxfendazole concentration in the specimen. As whole blood samples would be collected and centrifuged in the same clinical research unit in the Institute for Clinical and Translational Science, University of Iowa, we speculate that 4 hours would be more than sufficient for the study personnel to centrifuge whole blood samples after collection. Therefore, in this study, we have evaluated oxfendazole stability in whole blood for up to 4 hours. Blank whole blood donated by a healthy volunteer at the University of Iowa Hospital and Clinics was collected in a test tube with sodium heparin as anticoagulant. Oxfendazole working solution prepared in sodium chloride 0.9% was spiked into whole blood to a final concentration of 75 ng/mL. After oxfendazole addition, whole blood sample was gently mixed for 5 minutes, and 100 uL aliquots were collected after 10 minutes, 1, 2, and 4 hours at room temperature. Whole blood samples were prepared and analyzed following the procedures described in section 2.4 and 2.2. Oxfendazole concentration at 1, 2, and 4 hours was compared with that at 10 minutes. A difference within ±15.0% and CV ≤15.0% were set as acceptance criteria for oxfendazole stability in whole blood with sodium heparin.

Relative drug distribution in plasma and red blood cells (RBCs) can be an important parameter in pharmacokinetics, pharmacodynamics, and toxicity. The concentration of drug in RBCs relative to that in plasma is represented by RBC-to-plasma ratio.

$$\text{RBC-to-plasma ratio:}{\text{K}}_{\frac{\text{RBC}}{\text{plasma}}}=\frac{1}{\text{H}}\times \left(\frac{{\text{I}}_{\text{REF,Pl}}}{{\text{I}}_{\text{Pl}}}-1\right)+1$$

Where H is hematocrit obtained by hematological analysis using Sysmex XE–2100 (Sysmex, Kobe, Japan), IREF,Pl and IPl are IS-normalized peak area of oxfendazole in reference plasma and in whole blood sample, respectively.

c). Stock solution stability

The stability of oxfendazole stock solution in methanol was evaluated at −20°C for 2 weeks and 2 months, and at 4°C for 3 months. Oxfendazole samples with final concentration of 1000 ng/mL in mobile phase were prepared in triplicate from old and freshly made stock solutions. Albendazole was added to each sample to obtain a final concentration of 250 ng/mL. The stability of oxfendazole was tested by comparing the instrument response of samples prepared from old stock solution with that of samples prepared from new stock solution. Oxfendazole was considered stable in methanol if the mean difference fell within ±15.0% and the CV is no more than 15.0%.

2.6. Method Application to Clinical Pharmacokinetic Study

The validated method was applied to quantify oxfendazole plasma concentration obtained from 36 participants in the multiple ascending dose study of oxfendazole in healthy adult volunteers (ClinicalTrials.gov Identifier: NCT03035760). This is an open label trial that aims 1) to assess oxfendazole pharmacokinetics and safety after multiple ascending doses of oxfendazole (3, 7.5 and 15 mg/kg), and 2) to evaluate food effect on oxfendazole pharmacokinetics and safety after a single oral dose at 3 mg/kg. In the multiple dose trial, 24 subjects were randomized into three dose groups and received oxfendazole oral suspension (5, 7.5, or 15 mg/kg) once daily for 5 consecutive days. The food effect trial is a cross-over study in which half of the subjects received a single oral dose of oxfendazole at 3 mg/kg following 8 hours fasting, and the remaining subjects received 3 mg/kg of oxfendazole orally after a high fat meal. After a washout period of 7 days, subjects were crossed over to receive a single dose of oxfendazole either after a high fat meal or 8 hours fasting. At predetermined time points, blood samples were collected into a sodium heparin tube, stored on ice for no more than 1 hour, and centrifuged at 3000 rpm and 4°C for 15 minutes. Plasma was separated and stored at −80°C until analysis.

3. Results and Discussion

3.1. Sample Preparation, Chromatography and MS/MS

To minimize sample turnover time, a simple sample preparation procedure is highly desirable. In our method, simple acetonitrile protein precipitation was used in sample preparation. Acetonitrile was chosen as the precipitant in our method because it is one of the most efficient protein precipitants, especially when acetonitrile to plasma volume ratio is 2:1 or higher [18]. Among organic solvents commonly used in LC/MS/MS analysis, acetonitrile has lowest ionization suppression [18]. Therefore, in the presented method, plasma was mixed with acetonitrile in 1 to 2 ratio, which allowed a simple and efficient protein extraction process with good analyte recovery.

Chromatography conditions were evaluated in order to achieve the best separation, peak shape, and retention for oxfendazole and albendazole while minimizing the total sample running time. Phenomenex Synergi™ Polar-RP LC column, which is an ether-linked phenyl base with polar endcapping, was chosen as the column for the tested analytes due to its robustness in providing excellent retention for both hydrophobic as well as polar compounds. For organic mobile phase, both methanol and acetonitrile (each containing 0.1% formic acid) were evaluated. Shifting oxfendazole retention time was observed when methanol was used as the organic phase. With acetonitrile as the organic phase, good peak shapes and consistent retention were obtained for both oxfendazole and albendazole. Furthermore, in comparison with methanol, acetonitrile resulted in a significant reduction in pump pressure. Therefore, acetonitrile was finally chosen as the organic phase. Gradient flow was not used in our method because isocratic elution at a flow rate of 0.2 mL/min provided sufficient peak separation with desirable total run time.

To identify the mass spectra of oxfendazole and albendazole, 1000 ng/mL of each analyte were prepared individually in mobile phase (water–acetonitrile (50:50, v/v)) and injected into LC/MS/MS under positive ionization mode with a 0.2 mL/min flow rate. The mass spectra of oxfendazole, albendazole, and their product ions are illustrated in Figure 1. Precursor ions of oxfendazole and albendazole have m/z values of 316.100 and 266.300, respectively, which correspond to the protonated molecules ([M+H]⁺). After collision induced dissociation, the most abundant fragment of oxfendazole has m/z 191.300, probably due to sulfur α–cleavage, and the dominating fragment of albendazole has m/z 234.100, most likely due to CH₃OH loss. All source- and compound-dependent MS/MS parameters were finalized using flow injection analysis (FIA), a built-in function for automatic optimization in the Analyst software.

Figure 1.

Mass spectra of oxfendazole and albendazole: the injection of 1000 ng/mL of oxfendazole and albendazole separately with mass full scan. Oxfendazole exhibited the precursor ion of m/z 316.1 (A) and the most abundant product ion of m/z 191.3 (B). Albendazole exhibited the precursor ion of m/z 266.3 (C) and the most abundant resulting product ion has m/z 234.1(D).

3.2. Matrix Selectivity, Sensitivity, Linearity, and Injection Carryover

Under the optimized conditions, no visible interference at the elution times of oxfendazole and albendazole was observed in the blank and zero samples (Figure 2a and 2b). LLOQ of the presented method was 0.5 ng/mL with signal-to-noise ratio of 20:1 (Figure 2c), and accuracies and precisions within 20.0% (Table 1). The linear range was 0.5 – 1000 ng/mL. A typical linear equation for the calibration curve is y = 0.000656x + 0.000103. No carryover was detected in blank sample injected after a ULOQ sample.

Figure 2.

Chromatography of oxfendazole (left) and albendazole (right) in (A) blank sample; (B) zero sample; (C) oxfendazole LLOQ standard sample; and (D) oxfendazole ULOQ standard sample.

Table 1.

Intra-day and inter-day precision and accuracy, and reinjection reproducibility of oxfendazole in human plasma.

Concentration (ng/mL)	Intra-day (N=6)			Inter-day (N = 18)			Reinjection reproducibility (N = 6)
	Mean ± SD (ng/mL)	Accuracy (Bias, %)	Precision (CV, %)	Mean ± SD (ng/mL)	Accuracy (Bias, %)	Precision (CV, %)	Mean ± SD (ng/mL)	Accuracy (Bias, %)	Precision (CV, %)
0.5 (LLOQ)	0.513 ± 0.025	2.6	4.8	0.543 ± 0.082	8.6	15.1	0.444 ± 0.042	−11.3	9.4
1.5 (QC low)	1.62 ± 0.22	8.0	13.6	1.60 ± 0.19	6.9	11.8	1.36 ± 0.10	−9.7	7.4
75 (QC med)	82.1 ± 4.2	9.5	5.1	81.1 ± 3.6	8.1	4.4	80.9 ± 6.0	7.9	7.4
750 (QC high)	819 ± 32	9.2	4.0	815 ± 41	8.6	5.0	804 ± 23	7.2	2.8

3.3. Intra-day and Inter-day Precision and Accuracy, and Reinjection Reproducibility

Table 1 presents a summary of inter- and intra-day accuracies and precisions, and reinjection of oxfendazole samples in human plasma with sodium heparin. For QC low, medium, and high, the mean accuracies of the inter-day QC samples ranged from 106.9 to 108.6%, with CVs ranging from 4.4 to 11.8%. The mean intra-day accuracies of QC samples ranged from 108.0 to 109.5%. The CVs for all intra-day QC samples ranged from 4.0 to 13.6%. Linearity, accuracy, and precision were reproducible when samples were reinjected after 24 hours of storage in autosampler at 4°C.

3.4. Stability

Oxfendazole bench-top stability and long-term stability has been previously evaluated using a different method (manuscript to be submitted) and oxfendazole was found to be stable in human plasma at room temperature for at least 24 hours and at −80°C for 1 year. As these stability parameters are method-independent, they were not re-evaluated by the current method. In the current study, oxfendazole was confirmed to be stable in human plasma after 5 freeze/thaw cycles with CV less than 15.0% (Table 2).

Table 2.

Hemolysis effect and freeze/thaw stability (N = 6)

Concentration (ng/mL)	OXF in hemolyzed plasma			Freeze-thaw stability
	Mean ± SD (ng/mL)	Bias (%)	Cv (%)	Mean ± SD (ng/mL)	Bias (%)	CV (%)
1.5 (QC low)	1.54 ± 0.08	2.4	5.4	1.37 ± 0.20	−8.8	14.6
750 (QC high)	826 ± 12	10.1	1.5	670 ± 69	−10.7	10.3

Oxfendazole stock solution in methanol is stable at −20°C for 2 months and stable at 4°C for 3 months with accuracies ranged from −9.4 to 8.3% and CV% ≤4.9%. Oxfendazole is stable in whole blood at room temperature for 4 hours with CV of 5.2% (Table 3). In this experiment, RBC-to-plasma partition of oxfendazole was estimated to be 0.49, implying preferential partitioning of oxfendazole to plasma.

Table 3.

Oxfendazole whole blood stability (N = 6).

Time	IS-normalized peak area of oxfendazole in whole blood
	Mean ± SD	Bias (%)	CV (%)
10 minutes*	0.407 ± 0.007	0	1.8
1 hour	0.405 ± 0.013	−0.5	3.1
2 hours	0.379 ± 0.014	−6.9	3.6
4 hours	0.403 ± 0.021	−1.0	5.2

^*An outlier was excluded due to the use of wrong oxfendazole working solution.

3.5. Other Parameters

Matrix effect and extraction recovery evaluation are summarized in Table 4. Mean extraction recovery was in the range of 71.2 – 98.5%. The higher recovery observed at higher oxfendazole concentration was probably due to the analyte’s hydrophobicity. Matrix effect evaluated at 3 concentration levels (QC low, QC medium, and QC high) varied from 64.6 to 93.0%, suggesting a suppression effect in the presence of matrix. Despite a change in extraction recovery and matrix effect with oxfendazole concentration across different QC levels, extraction recovery and matrix effect tests were considered passed in our study as the only criteria for extraction recovery and matrix effect, according to the most recent FDA guidance, is that CV must be less than 15.0%.

Table 4.

Matrix effect and extraction recovery of oxfendazole in human plasma (N = 6).

Concentration (ng/mL)	Matrix effect		Extraction recovery
	Mean ± SD (%)	CV (%)	Mean ± SD (%)	CV (%)
1.5 (QC low)	64.6 ±6.5	10.1	98.5 ± 10.4	10.5
75 (QC med)	80.1 ± 11.8	14.8	86.4 ± 2.11	2.4
750 (QC high)	93.0 ± 12.4	13.3	71.2 ± 7.7	10.8

Hemolysis, defined as the rupture of red blood cells, can affect sample matrix or can release factors that may interfere with analyte detection, thus influencing analytical quantification of analyte in clinical samples. A comparison between oxfendazole detection in 2% hemolyzed plasma and in blank plasma at QC high and QC low shown that 2% hemolysis had little effect on oxfendazole quantification. Measured oxfendazole concentrations were within 102.4 –110.1% of nominal concentration and CVs were in the range of 1.5 –5.4% (Table 2). It is important to note that the significance of hemolysis effect might vary depending on the degree of hemolysis in a sample.

Mean albendazole peak area in interference samples was 0.17% of albendazole peak area detected in calibration standards and QC samples. Thus, there is no detectable interference of oxfendazole on IS.

Evaluation of oxfendazole concentration in six 15000 ng/mL samples after 15–fold dilution with blank human plasma resulted in a mean accuracy of 108.5% and CV of 2.7%. Thus, the developed assay can be used to accurately quantify oxfendazole in human plasma samples that are diluted to a maximum of 15 folds.

3.6. Method Application to Clinical Pharmacokinetic Study

The validated LC/MS/MS method was successfully applied to measure oxfendazole concentration in clinical samples obtained from the multiple ascending dose study and cross-over food effect study described in section 2.6. In general, after appropriate dilution of at most 15 folds, oxfendazole concentration in all clinical samples fall in the range of 0.647 – 1000 ng/mL. The pharmacokinetics results of oxfendazole in all 36 healthy volunteers in this clinical study will be presented in a separate manuscript. Here we only showed oxfendazole plasma concentration – time profile in one subject following the oral administration of multiple doses of oxfendazole at 3 mg/kg (Figure 3). Data for this one subject was analyzed by visual data inspection and non-compartmental analysis performed using Phoenix WinNonlin (Version 8.0, Certara). Time to peak concentration (Tmax) was 3.65 hours after first dose, and 2.53 hours after the last dose. Elimination half-life (t_1/2) was approximately 9.07 hours. The estimated Tmax and t_1/2 were within the range of Tmax and t_1/2 reported in the first-in-human study [16].

Figure 3.

Oxfendazole plasma concentration-time profile in an adult volunteer following 3 mg/kg oxfendazole doses once daily for 5 days

In general, the LC/MS/MS method presented in this study enabled accurate quantification oxfendazole in human plasma over a broad concentration range. On the lower end, we could measure oxfendazole in samples collected 5 days (i.e. >10 elimination half-lives) after the fifth dose even in the lowest dose group (i.e. 3 mg/kg). On the higher end, with 15-fold dilution, the developed LC/MS/MS method was able to capture all the peak plasma concentrations in all dose groups.

4. Conclusions

In this study, we have successfully developed a highly sensitive and efficient LC/MS/MS method for the detection of oxfendazole in human plasma using simple protein precipitation for sample preparation. Specifically, the total run time for each sample was 6 minutes and LLOQ was 0.5 ng/mL. This method was fully validated following FDA guidance for sensitivity, linearity, accuracy and precision, injection carryover, reproducibility, matrix effect, extraction recovery, interference of the analyte with IS, hemolysis effect, and dilution integrity. In addition, oxfendazole stock solution in methanol was confirmed to be stable at 4°C for 3 months and oxfendazole is stable in plasma after 5 freeze/thaw cycles. The method can be successfully employed in clinical studies investigating the potential transition of oxfendazole from animals to humans.

Highlights.

• Oxfendazole represents an attractive anthelminthic candidate for transition from veterinary use to human use. However, no analytical method has currently been published to detect oxfendazole in human matrix.

Highly sensitive LC-MS/MS method of oxfendazole in human plasma was established.

Full validations, accuracy and precision parameters all passed the FDA criteria.

The developed LC-MS/MS method of oxfendazole was successfully applied to a clinical pharmacokinetic study.

Abbreviations:

FIH

first-in-human

LLOQ

lower limit of quantitation

QC

quality control

IS

internal standard

ULOQ

upper limit of quantitation

CV

coefficient of variation

MRM

multiple reaction monitoring