Ultra-fast quantitation of voriconazole in human plasma by coated blade spray mass spectrometry

Highlights

• A Coated Blade Spray method for voriconazole determination from human plasma was developed.

• Analysis from from spots (10 μL) and small volumes (300 μL) of plasma were performed.

• Total analysis times obtained were between 120 and 180 s.

• The methodology was successfully validated for different plasma lots and patients.

• Relative matrix effects as low as 7% were achieved.

Abstract

Voriconazole is a triazole broad-spectrum antifungal medication often used to treat fungal infections caused by Aspergillus and Fusarium species. One of the main challenges associated with the implementation of this medication is its narrow therapeutic concentration range, demonstrating toxicity at concentrations above 6 μg/mL and limited efficacy at concentrations below 2 μg/mL. As a result, methodologies which permit the rapid and accurate quantitation of voriconazole in patients are highly desirable. In this work two different approaches based on coated blade spray directly coupled to mass spectrometry (CBS-MS) are introduced; each enabling the quantitation of voriconazole in plasma samples with a simple and fast sample preparation and no chromatographic step. The first approach involves a rapid extraction (1 min) of the target analyte from 300 μL of human plasma using conventional laboratory vessels (e.g. vial, 96-well plate). Alternatively, the second strategy consists of a 2 min extraction from a plasma droplet (10 μL) placed on the coated area of the blade. Both procedures were successfully validated and good linearity (R² ≥ 0.998), accuracy (91–122%) and precision (<8%) were attained in the concentration range evaluated (0.1–50 μg/mL). Moreover, very good results in terms of relative matrix effects were obtained given that the slopes of the calibration curves constructed in five different plasma lots exhibited relative standard deviation (RSD) values below 7%. Herein we demonstrated that CBS-MS is a technology suitable for the ultra-fast determination of voriconazole in human plasma samples. Indeed, the proposed methodology can be easily used either for routine drug monitoring or for in vitro pharmacokinetic studies in applications where very small sample volumes are available and great temporal resolution is needed.

Introduction

Voriconazole is an antifungal drug from the family of the triazoles. As a derivative of fluconazole, voriconazole possesses a fluoropyrimidine group rather than a triazole moiety in addition to an alpha methylation on the tertiary carbon [1]. In contrast to the first generation of triazole antifungals, voriconazole has a broader spectrum of action [2], [3] and has been demonstrated as effective against fungal infections caused by Aspergillus [4] and Fusarium species [3]. The mechanism of action consists of the inhibition of cytochrome P450 enzyme lanosterol 14-alpha-demethylase, which prevents the conversion of lanosterol to ergosterol, an essential component of cell membranes [5]. Although the effectiveness of this antifungal drug has been widely proven, toxicity risks and a narrow therapeutic concentration window have been reported. Indeed, plasma concentrations above 6 μg/mL have been associated with clinical events such as visual impairment or photopsia, abnormal hepatic function, and higher bilirubin levels [6], [7]. On the other hand, in cases where voriconazole plasma concentrations were below 2 μg/mL, a great proportion of patients suffered a significant progression in their infections [8]. In addition, high pharmacokinetic variability of this drug due to erratic bioavailability and variation in drug metabolism has also been documented [9]. Consequently, the development of an accurate, reliable, fast, and cost-effective analysis of voriconazole in plasma is highly desirable.

Most of the methods for voriconazole determination to date involve the use of liquid chromatography (LC) with either ultra-violet (UV) [1] or mass spectrometry (MS) [10], [11], [12], [13] detection. In addition, methods employing gas chromatography–MS (GC–MS) [14] and capillary electrophoresis (CE) [15] have also been reported. Within the LC–MS applications, the fastest LC separation was attained in 3 min [16] and the sample volume consumption was in the range between 20 μL [10] and 200 μL [16] of plasma. However, when the sample preparation step is considered in the total analysis time, none of the published methodologies can be completed in less than 15 min considering that even in the simplest approaches protein precipitation and centrifugation steps are required in order to have an extract suitable for liquid chromatography [12]. Recent developments in bioanalytical applications using solid phase microextraction (SPME) have demonstrated that this technology is an attractive alternative to conventional methods due to its simplicity, sensitivity and speed of analysis [17]. In SPME, a device coated with a polymeric material is used to extract/enrich analytes from a given sample prior to instrumental analysis. By using novel ultra-thin biocompatible coatings, faster and efficient extractions of small molecules from biofluids can be easily attained [18], [19], [20]. Despite the well-known advantages of using LC, emerging technologies such ambient mass spectrometry [21], [22], [23], [24], [25] and the direct coupling of SPME to very sensitive and selective mass spectrometers have shown outstanding results when aiming to decrease the total analysis time [26]. Among the recently developed SPME-MS technologies, coated blade spray (CBS) [17], SPME-direct analysis in real time (SPME-DART) [27], Bio-SPME-nano-ESI [26], Bio-SPME-OPP [28] and SPME-DBDI [29] excel by the limits of detection achieved. Succinctly, CBS was developed as an ideal compromise between sample preparation and direct coupling to mass spectrometry and it behaves as an SPME device for extraction and as a solid-substrate ESI source when performing ionization. When comparing CBS to other SPME-MS technologies, the former offers additional advantages such as simplicity of the set-up (see Fig. 1), negligible MS carry over (i.e. no-cross talking among experiments), minimal solvent use (≤20 μL), and no-need of expensive parts for the construction of the ionization source [17]. In addition, CBS is the device with the largest coated surface area available for extraction when compare to the other SPME-devices used for direct coupling to MS; therefore, low or even sub parts per billion levels can be easily achieved when rapid extractions (time ≤ 2 min) are performed from small sample volumes (≤300 μL).

In this work, we propose the ultra-fast determination of voriconazole in human plasma by two different CBS-based approaches. In the first strategy, voriconazole is extracted on the CBS device by fully immersion of the coated blade in 300 μL of human plasma during 1 min under vortex agitation. In the second approach, analytes are enriched on the CBS by placing a 10 μL droplet of plasma for 2 min on the coated area. After the completion of the extraction process in both cases, a quick rinsing step is performed in order to remove matrix components potentially adhered to the surface. Once complete, analytes were directly desorbed/ionized from the CBS device and analyzed by mass spectrometry without the need of additional sample preparation or chromatography. Both methodologies were fully validated in terms of linearity, LOD, LOQ, repeatability, reproducibility, relative matrix effects and accuracy employing 5 different lots of plasma and 4 different patients.