Determination of Prostaglandin E₁ in dog plasma using liquid chromatography–tandem mass spectrometry and its application to a pharmacokinetic study

Highlights

• An LC–MS/MS was developed and validated for the determination of PGE1 in beagle dog plasma.

• An LLOQ of 10 pg/mL was achieved.

• Indomethacin was added to plasma to avoid significant endogenous interfere.

• A simple one-step sample extraction was used followed by a short cycle time of 3 min.

• The method was applied to a pharmacokinetic study in dogs at a clinical relevant dose.

Abstract

The determination of Prostaglandin (PG) E₁ in plasma is challenged by its low concentration (pg/mL) and endogenous interference. An LC–MS/MS method for the determination of PGE₁ in dog plasma has been developed and validated. Plasma being sampled at 4 °C and treated with indomethacin effectively inhibited interferents synthesized post-sampling. Samples were subjected to one-step extraction and separated by reversed phase HPLC with a short cycle time of 3 min. An LLOQ of 10 pg/mL was achieved with 500 μl plasma. The method was applied to a pharmacokinetic study in beagle dogs involving an intravenous infusion of 3.2 μg/kg PGE₁. The half-life was recovered at 7 min. The simple, sensitive and rapid method was suitable to be applied to pharmacokinetic studies of PGE₁ at clinically relevant doses.

Introduction

Prostaglandin (PG) E₁ (Fig. 1), also known as alprostadil, is a vasodilator widely used for the treatment of ischemic peripheral vascular disease [1], [2], [3], [4]. It exerts multiple effects on peripheral vascular system, including increase of peripheral blood flow [5] and viscosity [6], modulation of fibrinolytic system [7], and inhibition of platelet aggregation [8]. The drug is also administrated intraurethrally and intracavernosally as a second-line therapy for erectile dysfunction [9]. It has been known that PGE₁ is inactivated rapidly by lung [10], [11]. Based on its short half-life, the drug is recommended to be administrated by intravenous infusion to maintain plasma PGE₁ concentrations and thus prolong its action for the treatment of peripheral arterial occlusive disease [12].

Recently, prolonged release preparations of PGE₁ were developed in order to expand its administration routes and facilitate clinical application. However, it was a challenge to evaluate these preparations by pharmacokinetic studies due to the difficulty to determine PGE₁ in plasma as well as in other biological samples. Furthermore, although the drug has been applied in clinical practice for decades, to date there are very few reports on its pharmacokinetics. The determination of plasma PGE₁ is not only challenged by its low levels (pg/mL) following a dose in microgram order [13], but also challenged by the endogenous interferents. Endogenous PGE₁ at normal physiological level is very low in plasma, such as 1–3 pg/mL in human plasma [14], and therefore does not significantly interference the determination of exogenous PGE₁. However, great amounts of interferents, probably PGE₁ and other PGs, could be biosynthesized from the phospholipids released from cell membrane residues remaining in plasma. The occurrence of post-sampling interference at high levels made it impossible to quantitation exogenous PGE₁ in low pg/mL order.

Several methods have been reported concerned with the determination of PGE₁ in biological matrix. Schweer et al. [14] and Hammes et al. [15] reported gas chromatography-tandem mass spectrometry (GC-MS/MS) methods for the determination of PGE₁ in human plasma, with a lower limit of detection (LLOD) of 1 pg/mL and a lower limit of quantitation (LLOQ) of 2 pg/mL, respectively. Although both of the methods provided high sensitivity, they suffered from time-consuming sample preparation procedure of three step derivatization. High performance liquid chromatography (HPLC) based methods were developed by Hotter et al. [16] and Tsutsumiuchi et al. [17] with LLOQs of 3.9 pg/mL and 23 pg/mL, respectively. Expect for analytical run time (30 min), long time was consumed on either radioimmunoassay (24 h) or derivatization (1 h). None of the methods discussed above was suitable to be applied to pharmacokinetic studies based on their low throughputs, e.g. only 24 samples could be prepared per day in the GC-MS/MS method with LLOQ of 2 pg/mL [15]. Rapid and simple LC–MS methods were provided by Abián et al. [18] and Lee et al. [19] with LLOQs of 50 ng/mL–20 μg/mL. Lin et al. reported an ultra-high performance liquid chromatography–tandem mass spectrometry (UPLC–MS/MS) method for the determination of PGE₁ in plasma and the method was applied to a pharmacokinetic study involving an intravenous injection of PGE₁ [20]. The method was rapid and simple with one-step sample preparation and run time of 2 min. However, it suffered from an LLOQ of 400 pg/mL, which led to a non-clinically-relevant dose of 50 μg/kg PGE₁ for the pharmacokinetic study in rats.

In order to solve these problems, a liquid chromatography–tandem mass spectrometric (LC–MS/MS) method for the determination of PGE₁ in dog plasma was developed and it has been successfully applied to a pharmacokinetic study in beagle dogs involving an intravenous infusion of PGE₁ at a therapeutically relevant dose of 3.2 μg/kg. The method involved one-step sample preparation, a short analytical time (cycle time 3 min) and high sensitivity (LLOQ 10 pg/mL). Indomethacin was added to plasma to inhibit post-sampling synthesis of endogenous interferences and low baselines were obtained. The sensitivity of this method (LLOQ 10 pg/mL using 0.5 mL plasma) was comparable with the GC–MS/MS method with an LLOQ of 2 pg/mL using 2 mL plasma, but the former method was much more simple and rapid. This method is being validated with human plasma and will be applied to a phase I clinical study for PGE₁ micelles. An LLOQ of 1 pg/mL has been achieved using 2 mL human plasma.